Apple M5 Max (64 GB) vs AMD Radeon PRO W7800 48GB for local AI
Pick two GPUs or Macs to see which runs more models, faster, for the kind of work you do.
Too close to call
For general chat, the Apple M5 Max (64 GB) and AMD Radeon PRO W7800 48GB come out about even. Look at price and the specific models you care about below.
Side by side
| Apple M5 Max (64 GB) | AMD Radeon PRO W7800 48GB | |
|---|---|---|
| Memory | 64.0 GB unified | 48.0 GB VRAM |
| Usable for models | 48.0 GB | 48.0 GB |
| Memory bandwidth | 614 GB/s | 864 GB/s |
| FP16 compute | 18.9 TFLOPS | 90.4 TFLOPS |
| Launch price | — | $2,799 |
| Released | 2026 | 2024 |
| LLMs that run | 459 of 536 | 461 of 536 |
| LLMs that run well | 439 | 443 |
| Image models that run | 47 of 47 | 47 of 47 |
| Video models that run | 34 of 34 | 34 of 34 |
| Top-10 LLM average score | 99 | 99 |
| Top-10 LLM average speed | 71.0 tok/s | 94.2 tok/s |
| Best LLM | Qwen3.5 35B-A3B | Ornith 1.5 35B-A3B |
Highlighted values are the better of the two. Model results assume 8K of context and the default macOS GPU memory limit on Macs.
LLMs
| Model | Grade · points (Apple M5 Max (64 GB)) | Speed (Apple M5 Max (64 GB)) | Fit (Apple M5 Max (64 GB)) | Grade · points (AMD Radeon PRO W7800 48GB) | Speed (AMD Radeon PRO W7800 48GB) | Fit (AMD Radeon PRO W7800 48GB) |
|---|---|---|---|---|---|---|
| Qwen3.5 35B-A3B | 100 ptsExcellent (S) · 100 of 100 points: Model quality 40.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 14.5/15, workload match +5.0 | 89.4 tok/s | Runs well | 99 ptsExcellent (S) · 99 of 100 points: Model quality 40.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 13.8/15, workload match +5.0 | 126 tok/s | Runs well |
| Qwen3.6 35B-A3B | 100 ptsExcellent (S) · 100 of 100 points: Model quality 40.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 14.5/15, workload match +5.0 | 89.4 tok/s | Runs well | 99 ptsExcellent (S) · 99 of 100 points: Model quality 40.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 13.8/15, workload match +5.0 | 126 tok/s | Runs well |
| Ornith 1.5 35B-A3B | 100 ptsExcellent (S) · 100 of 100 points: Model quality 40.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 75.1 tok/s | Runs well | 100 ptsExcellent (S) · 100 of 100 points: Model quality 40.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 106 tok/s | Runs well |
| Apodex 1.1 Mini (35B-A3B) | 99 ptsExcellent (S) · 99 of 100 points: Model quality 39.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 75.1 tok/s | Runs well | 99 ptsExcellent (S) · 99 of 100 points: Model quality 39.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 106 tok/s | Runs well |
| Ornith 1.0 35B-A3B | 99 ptsExcellent (S) · 99 of 100 points: Model quality 39.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 75.1 tok/s | Runs well | 99 ptsExcellent (S) · 99 of 100 points: Model quality 39.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 106 tok/s | Runs well |
| GLM-4.7 Flash (30B-A3B) | 99 ptsExcellent (S) · 99 of 100 points: Model quality 39.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 55.1 tok/s | Runs well | 99 ptsExcellent (S) · 99 of 100 points: Model quality 39.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 77.5 tok/s | Runs well |
| Agents-A1 35B-A3B | 98 ptsExcellent (S) · 98 of 100 points: Model quality 38.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 75.1 tok/s | Runs well | 99 ptsExcellent (S) · 99 of 100 points: Model quality 39.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 77.5 tok/s | Runs well |
| Qwen AgentWorld 35B-A3B | 98 ptsExcellent (S) · 98 of 100 points: Model quality 38.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 75.1 tok/s | Runs well | 99 ptsExcellent (S) · 99 of 100 points: Model quality 39.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 77.5 tok/s | Runs well |
| Gemma 4 26B-A4B | 98 ptsExcellent (S) · 98 of 100 points: Model quality 41.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 12.5/15, workload match +5.0 | 59.3 tok/s | Runs well | 98 ptsExcellent (S) · 98 of 100 points: Model quality 41.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 11.8/15, workload match +5.0 | 83.4 tok/s | Runs well |
| DiffusionGemma 26B-A4B IT | 95 ptsExcellent (S) · 95 of 100 points: Model quality 35.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 41.3 tok/s | Runs well | 94 ptsExcellent (S) · 94 of 100 points: Model quality 35.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 14.8/15, workload match +5.0 | 58.1 tok/s | Runs well |
| Nemotron 3 Nano 30B-A3B | 94 ptsExcellent (S) · 94 of 100 points: Model quality 34.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 47.2 tok/s | Runs well | 94 ptsExcellent (S) · 94 of 100 points: Model quality 34.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 66.4 tok/s | Runs well |
| Qwen3 30B-A3B Instruct 2507 | 94 ptsExcellent (S) · 94 of 100 points: Model quality 34.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 50.1 tok/s | Runs well | 94 ptsExcellent (S) · 94 of 100 points: Model quality 34.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 70.4 tok/s | Runs well |
| Qwen3-VL 30B-A3B Thinking | 94 ptsExcellent (S) · 94 of 100 points: Model quality 34.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 50.1 tok/s | Runs well | 94 ptsExcellent (S) · 94 of 100 points: Model quality 34.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 70.4 tok/s | Runs well |
| Qwen3 30B-A3B | 93 ptsExcellent (S) · 93 of 100 points: Model quality 33.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 50.1 tok/s | Runs well | 93 ptsExcellent (S) · 93 of 100 points: Model quality 33.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 70.4 tok/s | Runs well |
| Qwen3-VL 30B-A3B Instruct | 93 ptsExcellent (S) · 93 of 100 points: Model quality 33.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 50.1 tok/s | Runs well | 93 ptsExcellent (S) · 93 of 100 points: Model quality 33.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 70.4 tok/s | Runs well |
| Qwen3.8 27B | 92 ptsExcellent (S) · 92 of 100 points: Model quality 40.9/45, Memory fit 25.0/25, Speed 11.5/15, Memory use 10.5/15, workload match +5.0 | 23.4 tok/s | Runs well | 94 ptsExcellent (S) · 94 of 100 points: Model quality 40.9/45, Memory fit 25.0/25, Speed 13.6/15, Memory use 9.8/15, workload match +5.0 | 32.9 tok/s | Runs well |
| Nemotron 3.5 Lightning 30B-A3B | 92 ptsExcellent (S) · 92 of 100 points: Model quality 32.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 55.1 tok/s | Runs well | 92 ptsExcellent (S) · 92 of 100 points: Model quality 32.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 77.5 tok/s | Runs well |
| Kimi Linear 48B-A3B | 92 ptsExcellent (S) · 92 of 100 points: Model quality 32.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 89.4 tok/s | Runs well | 92 ptsExcellent (S) · 92 of 100 points: Model quality 32.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 126 tok/s | Runs well |
| Qwen3.6 27B | 92 ptsExcellent (S) · 92 of 100 points: Model quality 40.2/45, Memory fit 25.0/25, Speed 11.5/15, Memory use 10.5/15, workload match +5.0 | 23.4 tok/s | Runs well | 93 ptsExcellent (S) · 93 of 100 points: Model quality 40.2/45, Memory fit 25.0/25, Speed 13.6/15, Memory use 9.8/15, workload match +5.0 | 32.9 tok/s | Runs well |
| Gemma 4 31B | 91 ptsExcellent (S) · 91 of 100 points: Model quality 38.9/45, Memory fit 25.0/25, Speed 10.8/15, Memory use 12.1/15, workload match +5.0 | 20.8 tok/s | Runs well | 93 ptsExcellent (S) · 93 of 100 points: Model quality 38.9/45, Memory fit 25.0/25, Speed 12.8/15, Memory use 11.4/15, workload match +5.0 | 29.2 tok/s | Runs well |
| Qwen3.5 27B | 91 ptsExcellent (S) · 91 of 100 points: Model quality 39.8/45, Memory fit 25.0/25, Speed 11.5/15, Memory use 10.5/15, workload match +5.0 | 23.4 tok/s | Runs well | 93 ptsExcellent (S) · 93 of 100 points: Model quality 39.8/45, Memory fit 25.0/25, Speed 13.6/15, Memory use 9.8/15, workload match +5.0 | 32.9 tok/s | Runs well |
| Laguna XS 2.1 (33B-A3B) | 91 ptsExcellent (S) · 91 of 100 points: Model quality 36.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | 75.1 tok/s | Runs well | 91 ptsExcellent (S) · 91 of 100 points: Model quality 36.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | 77.5 tok/s | Runs well |
| Qwen3-Next 80B-A3B Thinking | 91 ptsExcellent (S) · 91 of 100 points: Model quality 32.5/45, Memory fit 24.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 140 tok/s | Tight fit | 91 ptsExcellent (S) · 91 of 100 points: Model quality 32.5/45, Memory fit 24.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 197 tok/s | Tight fit |
| Sarvam 30B | 91 ptsExcellent (S) · 91 of 100 points: Model quality 31.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 68.8 tok/s | Runs well | 91 ptsExcellent (S) · 91 of 100 points: Model quality 31.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 96.9 tok/s | Runs well |
| Qwen3-Omni 30B-A3B Instruct | 91 ptsExcellent (S) · 91 of 100 points: Model quality 31.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 68.2 tok/s | Runs well | 91 ptsExcellent (S) · 91 of 100 points: Model quality 31.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 96.0 tok/s | Runs well |
| Nemotron Cascade 2 30B-A3B | 91 ptsExcellent (S) · 91 of 100 points: Model quality 36.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | 47.2 tok/s | Runs well | 91 ptsExcellent (S) · 91 of 100 points: Model quality 36.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | 66.4 tok/s | Runs well |
| North Mini Code 1.0 (30B-A3B) | 91 ptsExcellent (S) · 91 of 100 points: Model quality 36.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | 55.1 tok/s | Runs well | 91 ptsExcellent (S) · 91 of 100 points: Model quality 36.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | 77.5 tok/s | Runs well |
| Qwen3 Next 80B-A3B Instruct | 90 ptsExcellent (S) · 90 of 100 points: Model quality 31.8/45, Memory fit 24.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 140 tok/s | Tight fit | 90 ptsExcellent (S) · 90 of 100 points: Model quality 31.8/45, Memory fit 24.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 197 tok/s | Tight fit |
| TMax 27B | 90 ptsExcellent (S) · 90 of 100 points: Model quality 38.5/45, Memory fit 25.0/25, Speed 11.6/15, Memory use 10.4/15, workload match +5.0 | 23.7 tok/s | Runs well | 91 ptsExcellent (S) · 91 of 100 points: Model quality 38.5/45, Memory fit 25.0/25, Speed 13.7/15, Memory use 9.7/15, workload match +5.0 | 33.4 tok/s | Runs well |
| Qwen3 30B-A3B Thinking 2507 | 90 ptsExcellent (S) · 90 of 100 points: Model quality 35.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | 50.1 tok/s | Runs well | 90 ptsExcellent (S) · 90 of 100 points: Model quality 35.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | 70.4 tok/s | Runs well |
| Qwen3 Coder 30B-A3B | 90 ptsExcellent (S) · 90 of 100 points: Model quality 35.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | 50.1 tok/s | Runs well | 90 ptsExcellent (S) · 90 of 100 points: Model quality 35.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | 70.4 tok/s | Runs well |
| Trinity Mini (26B-A3B) | 89 ptsExcellent (S) · 89 of 100 points: Model quality 29.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 55.1 tok/s | Runs well | 89 ptsExcellent (S) · 89 of 100 points: Model quality 29.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 14.8/15, workload match +5.0 | 77.5 tok/s | Runs well |
| Fara 1.5 27B | 89 ptsExcellent (S) · 89 of 100 points: Model quality 37.6/45, Memory fit 25.0/25, Speed 11.6/15, Memory use 10.4/15, workload match +5.0 | 23.7 tok/s | Runs well | 91 ptsExcellent (S) · 91 of 100 points: Model quality 37.6/45, Memory fit 25.0/25, Speed 13.7/15, Memory use 9.7/15, workload match +5.0 | 33.4 tok/s | Runs well |
| UI-Mate 27B | 88 ptsExcellent (S) · 88 of 100 points: Model quality 36.8/45, Memory fit 25.0/25, Speed 11.6/15, Memory use 10.4/15, workload match +5.0 | 23.7 tok/s | Runs well | 90 ptsExcellent (S) · 90 of 100 points: Model quality 36.8/45, Memory fit 25.0/25, Speed 13.7/15, Memory use 9.7/15, workload match +5.0 | 33.4 tok/s | Runs well |
| Gemma 4 12B | 88 ptsExcellent (S) · 88 of 100 points: Model quality 37.0/45, Memory fit 25.0/25, Speed 14.6/15, Memory use 7.2/15, workload match +5.0 | 37.7 tok/s | Runs well | 89 ptsExcellent (S) · 89 of 100 points: Model quality 37.4/45, Memory fit 25.0/25, Speed 14.8/15, Memory use 7.9/15, workload match +5.0 | 38.9 tok/s | Runs well |
| Seed-OSS 36B Instruct | 88 ptsExcellent (S) · 88 of 100 points: Model quality 34.2/45, Memory fit 25.0/25, Speed 10.0/15, Memory use 13.8/15, workload match +5.0 | 18.0 tok/s | Runs well | 89 ptsExcellent (S) · 89 of 100 points: Model quality 34.2/45, Memory fit 25.0/25, Speed 11.9/15, Memory use 13.1/15, workload match +5.0 | 25.3 tok/s | Runs well |
| EXAONE 4.5 33B | 87 ptsExcellent (S) · 87 of 100 points: Model quality 35.1/45, Memory fit 25.0/25, Speed 10.3/15, Memory use 12.6/15, workload match +5.0 | 18.9 tok/s | Runs well | 89 ptsExcellent (S) · 89 of 100 points: Model quality 35.1/45, Memory fit 25.0/25, Speed 12.2/15, Memory use 11.9/15, workload match +5.0 | 26.6 tok/s | Runs well |
| Llama 3.3 Nemotron Super 49B v1.5 | 87 ptsExcellent (S) · 87 of 100 points: Model quality 34.2/45, Memory fit 25.0/25, Speed 8.6/15, Memory use 15.0/15, workload match +5.0 | 13.0 tok/s | Runs well | 89 ptsExcellent (S) · 89 of 100 points: Model quality 34.2/45, Memory fit 25.0/25, Speed 10.1/15, Memory use 15.0/15, workload match +5.0 | 18.3 tok/s | Runs well |
| Hermes 4.3 36B | 87 ptsExcellent (S) · 87 of 100 points: Model quality 33.8/45, Memory fit 25.0/25, Speed 10.0/15, Memory use 13.8/15, workload match +5.0 | 18.0 tok/s | Runs well | 88 ptsExcellent (S) · 88 of 100 points: Model quality 33.8/45, Memory fit 25.0/25, Speed 11.9/15, Memory use 13.1/15, workload match +5.0 | 25.3 tok/s | Runs well |
| ERNIE 4.5 21B-A3B | 87 ptsExcellent (S) · 87 of 100 points: Model quality 28.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 13.5/15, workload match +5.0 | 55.1 tok/s | Runs well | 86 ptsExcellent (S) · 86 of 100 points: Model quality 28.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 12.9/15, workload match +5.0 | 77.5 tok/s | Runs well |
| Qwen3-VL 32B Instruct | 86 ptsExcellent (S) · 86 of 100 points: Model quality 33.3/45, Memory fit 25.0/25, Speed 10.5/15, Memory use 13.0/15, workload match +5.0 | 19.5 tok/s | Runs well | 88 ptsExcellent (S) · 88 of 100 points: Model quality 33.3/45, Memory fit 25.0/25, Speed 12.4/15, Memory use 12.3/15, workload match +5.0 | 27.4 tok/s | Runs well |
| InternVL3.5 38B | 86 ptsExcellent (S) · 86 of 100 points: Model quality 32.5/45, Memory fit 25.0/25, Speed 9.8/15, Memory use 14.4/15, workload match +5.0 | 16.9 tok/s | Runs well | 87 ptsExcellent (S) · 87 of 100 points: Model quality 32.5/45, Memory fit 25.0/25, Speed 11.6/15, Memory use 13.7/15, workload match +5.0 | 23.8 tok/s | Runs well |
| Qwen3 Coder Next (80B-A3B) | 86 ptsExcellent (S) · 86 of 100 points: Model quality 32.5/45, Memory fit 24.0/25, Speed 15.0/15, Memory use 15.0/15 | 140 tok/s | Tight fit | 86 ptsExcellent (S) · 86 of 100 points: Model quality 32.5/45, Memory fit 24.0/25, Speed 15.0/15, Memory use 15.0/15 | 197 tok/s | Tight fit |
| Qwen3 32B | 85 ptsExcellent (S) · 85 of 100 points: Model quality 32.5/45, Memory fit 25.0/25, Speed 10.6/15, Memory use 12.8/15, workload match +5.0 | 19.8 tok/s | Runs well | 87 ptsExcellent (S) · 87 of 100 points: Model quality 32.5/45, Memory fit 25.0/25, Speed 12.5/15, Memory use 12.1/15, workload match +5.0 | 27.8 tok/s | Runs well |
| Hunyuan A13B (80B-A13B) | 85 ptsExcellent (S) · 85 of 100 points: Model quality 28.0/45, Memory fit 24.0/25, Speed 13.5/15, Memory use 15.0/15, workload match +5.0 | 32.3 tok/s | Tight fit | 86 ptsExcellent (S) · 86 of 100 points: Model quality 28.0/45, Memory fit 24.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 45.4 tok/s | Tight fit |
| Granite 4.2 30B | 85 ptsExcellent (S) · 85 of 100 points: Model quality 32.5/45, Memory fit 25.0/25, Speed 11.2/15, Memory use 11.8/15, workload match +5.0 | 22.2 tok/s | Runs well | 86 ptsExcellent (S) · 86 of 100 points: Model quality 32.5/45, Memory fit 25.0/25, Speed 13.2/15, Memory use 11.1/15, workload match +5.0 | 31.2 tok/s | Runs well |
| Ornith 1.5 9B | 85 ptsExcellent (S) · 85 of 100 points: Model quality 34.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 46.7 tok/s | Runs well | 86 ptsExcellent (S) · 86 of 100 points: Model quality 34.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.5/15, workload match +5.0 | 48.2 tok/s | Runs well |
| ERNIE 4.5 21B-A3B Thinking | 84 ptsGreat (A) · 84 of 100 points: Model quality 31.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 13.5/15 | 55.1 tok/s | Runs well | 84 ptsGreat (A) · 84 of 100 points: Model quality 31.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 12.8/15 | 77.5 tok/s | Runs well |
| Qwen3.5 9B | 84 ptsGreat (A) · 84 of 100 points: Model quality 33.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 46.7 tok/s | Runs well | 85 ptsExcellent (S) · 85 of 100 points: Model quality 34.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.5/15, workload match +5.0 | 48.2 tok/s | Runs well |
| LFM2 24B-A2B | 84 ptsGreat (A) · 84 of 100 points: Model quality 25.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 14.4/15, workload match +5.0 | 71.8 tok/s | Runs well | 83 ptsGreat (A) · 83 of 100 points: Model quality 25.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 13.7/15, workload match +5.0 | 101 tok/s | Runs well |
| Ornith 1.0 9B | 84 ptsGreat (A) · 84 of 100 points: Model quality 33.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 46.7 tok/s | Runs well | 85 ptsExcellent (S) · 85 of 100 points: Model quality 33.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.5/15, workload match +5.0 | 48.2 tok/s | Runs well |
| gpt-oss 20B | 84 ptsGreat (A) · 84 of 100 points: Model quality 31.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.0/15, workload match +5.0 | 96.5 tok/s | Runs well | 83 ptsGreat (A) · 83 of 100 points: Model quality 31.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.3/15, workload match +5.0 | 136 tok/s | Runs well |
| Phi-3.5 MoE (42B-A6.6B) | 84 ptsGreat (A) · 84 of 100 points: Model quality 24.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 40.6 tok/s | Runs well | 84 ptsGreat (A) · 84 of 100 points: Model quality 24.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 48.0 tok/s | Runs well |
| Solar Open 100B | 84 ptsGreat (A) · 84 of 100 points: Model quality 25.2/45, Memory fit 24.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 44.4 tok/s | Tight fit | 84 ptsGreat (A) · 84 of 100 points: Model quality 25.2/45, Memory fit 24.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 62.5 tok/s | Tight fit |
| GLM-4 32B 0414 | 84 ptsGreat (A) · 84 of 100 points: Model quality 31.6/45, Memory fit 25.0/25, Speed 10.6/15, Memory use 11.9/15, workload match +5.0 | 19.9 tok/s | Runs well | 85 ptsExcellent (S) · 85 of 100 points: Model quality 31.6/45, Memory fit 25.0/25, Speed 12.6/15, Memory use 11.2/15, workload match +5.0 | 28.0 tok/s | Runs well |
| InternVL3 38B | 84 ptsGreat (A) · 84 of 100 points: Model quality 29.9/45, Memory fit 25.0/25, Speed 9.8/15, Memory use 14.4/15, workload match +5.0 | 16.9 tok/s | Runs well | 85 ptsExcellent (S) · 85 of 100 points: Model quality 29.9/45, Memory fit 25.0/25, Speed 11.6/15, Memory use 13.7/15, workload match +5.0 | 23.8 tok/s | Runs well |
| TMax 9B | 83 ptsGreat (A) · 83 of 100 points: Model quality 33.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.8/15, workload match +5.0 | 50.3 tok/s | Runs well | 84 ptsGreat (A) · 84 of 100 points: Model quality 33.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 51.9 tok/s | Runs well |
| Granite 4.1 30B | 83 ptsGreat (A) · 83 of 100 points: Model quality 30.8/45, Memory fit 25.0/25, Speed 11.2/15, Memory use 11.6/15, workload match +5.0 | 22.5 tok/s | Runs well | 85 ptsExcellent (S) · 85 of 100 points: Model quality 30.8/45, Memory fit 25.0/25, Speed 13.3/15, Memory use 10.9/15, workload match +5.0 | 31.6 tok/s | Runs well |
| Ling Mini 2.0 (16B-A1.4B) | 83 ptsGreat (A) · 83 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 54.4 tok/s | Runs well | 83 ptsGreat (A) · 83 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 76.5 tok/s | Runs well |
| Olmo 3.1 32B Instruct | 83 ptsGreat (A) · 83 of 100 points: Model quality 30.8/45, Memory fit 25.0/25, Speed 10.6/15, Memory use 11.9/15, workload match +5.0 | 20.1 tok/s | Runs well | 84 ptsGreat (A) · 84 of 100 points: Model quality 30.8/45, Memory fit 25.0/25, Speed 12.6/15, Memory use 11.2/15, workload match +5.0 | 28.3 tok/s | Runs well |
| EXAONE 4.0 32B | 83 ptsGreat (A) · 83 of 100 points: Model quality 30.8/45, Memory fit 25.0/25, Speed 10.7/15, Memory use 11.9/15, workload match +5.0 | 20.3 tok/s | Runs well | 84 ptsGreat (A) · 84 of 100 points: Model quality 30.8/45, Memory fit 25.0/25, Speed 12.7/15, Memory use 11.2/15, workload match +5.0 | 28.5 tok/s | Runs well |
| Qwen2.5 32B Instruct | 83 ptsGreat (A) · 83 of 100 points: Model quality 29.9/45, Memory fit 25.0/25, Speed 10.6/15, Memory use 12.8/15, workload match +5.0 | 19.8 tok/s | Runs well | 84 ptsGreat (A) · 84 of 100 points: Model quality 29.9/45, Memory fit 25.0/25, Speed 12.5/15, Memory use 12.1/15, workload match +5.0 | 27.8 tok/s | Runs well |
| Baichuan M2 32B | 83 ptsGreat (A) · 83 of 100 points: Model quality 29.9/45, Memory fit 25.0/25, Speed 10.6/15, Memory use 12.8/15, workload match +5.0 | 19.8 tok/s | Runs well | 84 ptsGreat (A) · 84 of 100 points: Model quality 29.9/45, Memory fit 25.0/25, Speed 12.5/15, Memory use 12.1/15, workload match +5.0 | 27.9 tok/s | Runs well |
| Llama 3.3 70B Instruct | 83 ptsGreat (A) · 83 of 100 points: Model quality 31.0/45, Memory fit 24.0/25, Speed 8.2/15, Memory use 15.0/15, workload match +5.0 | 11.9 tok/s | Tight fit | 85 ptsExcellent (S) · 85 of 100 points: Model quality 31.0/45, Memory fit 25.0/25, Speed 9.7/15, Memory use 15.0/15, workload match +5.0 | 16.7 tok/s | Runs well |
| Apriel 1.5 15B Thinker | 83 ptsGreat (A) · 83 of 100 points: Model quality 30.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.3/15, workload match +5.0 | 43.7 tok/s | Runs well | 85 ptsExcellent (S) · 85 of 100 points: Model quality 32.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.1/15, workload match +5.0 | 42.6 tok/s | Runs well |
| Ling Flash 2.0 (100B-A6B) | 82 ptsGreat (A) · 82 of 100 points: Model quality 23.9/45, Memory fit 24.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 87.4 tok/s | Tight fit | 82 ptsGreat (A) · 82 of 100 points: Model quality 23.9/45, Memory fit 24.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 123 tok/s | Tight fit |
| Gemma 3 27B IT | 82 ptsGreat (A) · 82 of 100 points: Model quality 30.8/45, Memory fit 25.0/25, Speed 11.5/15, Memory use 10.6/15, workload match +5.0 | 23.7 tok/s | Runs well | 84 ptsGreat (A) · 84 of 100 points: Model quality 30.8/45, Memory fit 25.0/25, Speed 13.7/15, Memory use 9.9/15, workload match +5.0 | 33.3 tok/s | Runs well |
| Sarvam 105B | 82 ptsGreat (A) · 82 of 100 points: Model quality 23.6/45, Memory fit 24.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 53.3 tok/s | Tight fit | 82 ptsGreat (A) · 82 of 100 points: Model quality 23.6/45, Memory fit 24.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 75.0 tok/s | Tight fit |
| Kimi-VL A3B Thinking 2506 | 82 ptsGreat (A) · 82 of 100 points: Model quality 27.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 10.6/15, workload match +5.0 | 55.1 tok/s | Runs well | 81 ptsGreat (A) · 81 of 100 points: Model quality 27.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 9.9/15, workload match +5.0 | 77.5 tok/s | Runs well |
| Qwen2.5-VL 32B Instruct | 82 ptsGreat (A) · 82 of 100 points: Model quality 29.1/45, Memory fit 25.0/25, Speed 10.4/15, Memory use 13.0/15, workload match +5.0 | 19.4 tok/s | Runs well | 83 ptsGreat (A) · 83 of 100 points: Model quality 29.1/45, Memory fit 25.0/25, Speed 12.4/15, Memory use 12.3/15, workload match +5.0 | 27.3 tok/s | Runs well |
| Granite 4.0 H Small (32B-A9B) | 82 ptsGreat (A) · 82 of 100 points: Model quality 26.5/45, Memory fit 25.0/25, Speed 14.2/15, Memory use 11.6/15, workload match +5.0 | 36.1 tok/s | Runs well | 84 ptsGreat (A) · 84 of 100 points: Model quality 27.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 12.5/15, workload match +5.0 | 41.9 tok/s | Runs well |
| Phi-4 Reasoning Vision 15B | 82 ptsGreat (A) · 82 of 100 points: Model quality 29.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.4/15, workload match +5.0 | 42.9 tok/s | Runs well | 84 ptsGreat (A) · 84 of 100 points: Model quality 31.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.2/15, workload match +5.0 | 41.9 tok/s | Runs well |
| Magistral Small 2507 | 82 ptsGreat (A) · 82 of 100 points: Model quality 29.9/45, Memory fit 25.0/25, Speed 12.4/15, Memory use 9.7/15, workload match +5.0 | 27.5 tok/s | Runs well | 83 ptsGreat (A) · 83 of 100 points: Model quality 29.9/45, Memory fit 25.0/25, Speed 14.8/15, Memory use 9.0/15, workload match +5.0 | 38.7 tok/s | Runs well |
| MedGemma 27B Text IT | 81 ptsGreat (A) · 81 of 100 points: Model quality 29.9/45, Memory fit 25.0/25, Speed 11.6/15, Memory use 10.4/15, workload match +5.0 | 24.0 tok/s | Runs well | 83 ptsGreat (A) · 83 of 100 points: Model quality 29.9/45, Memory fit 25.0/25, Speed 13.8/15, Memory use 9.8/15, workload match +5.0 | 33.8 tok/s | Runs well |
| Gemma 4 E4B | 81 ptsGreat (A) · 81 of 100 points: Model quality 30.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15, workload match +5.0 | 41.3 tok/s | Runs well | 81 ptsGreat (A) · 81 of 100 points: Model quality 30.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15, workload match +5.0 | 58.1 tok/s | Runs well |
| Ling 3.0 Tiny (8B-A1.3B) | 81 ptsGreat (A) · 81 of 100 points: Model quality 27.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 9.8/15, workload match +5.0 | 58.6 tok/s | Runs well | 81 ptsGreat (A) · 81 of 100 points: Model quality 27.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 9.1/15, workload match +5.0 | 82.4 tok/s | Runs well |
| QwQ 32B | 81 ptsGreat (A) · 81 of 100 points: Model quality 33.3/45, Memory fit 25.0/25, Speed 10.6/15, Memory use 12.8/15 | 19.8 tok/s | Runs well | 82 ptsGreat (A) · 82 of 100 points: Model quality 33.3/45, Memory fit 25.0/25, Speed 12.5/15, Memory use 12.1/15 | 27.9 tok/s | Runs well |
| Granite 4.2 8B | 81 ptsGreat (A) · 81 of 100 points: Model quality 30.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15, workload match +5.0 | 51.2 tok/s | Runs well | 82 ptsGreat (A) · 82 of 100 points: Model quality 30.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15, workload match +5.0 | 52.9 tok/s | Runs well |
| Mistral Small 3.2 24B | 81 ptsGreat (A) · 81 of 100 points: Model quality 29.1/45, Memory fit 25.0/25, Speed 12.3/15, Memory use 9.8/15, workload match +5.0 | 27.0 tok/s | Runs well | 82 ptsGreat (A) · 82 of 100 points: Model quality 29.1/45, Memory fit 25.0/25, Speed 14.6/15, Memory use 9.1/15, workload match +5.0 | 38.0 tok/s | Runs well |
| Magistral Small 2506 | 81 ptsGreat (A) · 81 of 100 points: Model quality 29.1/45, Memory fit 25.0/25, Speed 12.4/15, Memory use 9.7/15, workload match +5.0 | 27.5 tok/s | Runs well | 82 ptsGreat (A) · 82 of 100 points: Model quality 29.1/45, Memory fit 25.0/25, Speed 14.8/15, Memory use 9.0/15, workload match +5.0 | 38.7 tok/s | Runs well |
| Qwen3 14B | 81 ptsGreat (A) · 81 of 100 points: Model quality 29.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.1/15, workload match +5.0 | 43.9 tok/s | Runs well | 83 ptsGreat (A) · 83 of 100 points: Model quality 30.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.9/15, workload match +5.0 | 42.8 tok/s | Runs well |
| OpenReasoning Nemotron 32B | 80 ptsGreat (A) · 80 of 100 points: Model quality 32.5/45, Memory fit 25.0/25, Speed 10.6/15, Memory use 12.8/15 | 19.8 tok/s | Runs well | 82 ptsGreat (A) · 82 of 100 points: Model quality 32.5/45, Memory fit 25.0/25, Speed 12.5/15, Memory use 12.1/15 | 27.9 tok/s | Runs well |
| Skywork OR1 32B | 80 ptsGreat (A) · 80 of 100 points: Model quality 32.5/45, Memory fit 25.0/25, Speed 10.6/15, Memory use 12.8/15 | 19.8 tok/s | Runs well | 82 ptsGreat (A) · 82 of 100 points: Model quality 32.5/45, Memory fit 25.0/25, Speed 12.5/15, Memory use 12.1/15 | 27.9 tok/s | Runs well |
| Mistral Small 3.1 24B | 80 ptsGreat (A) · 80 of 100 points: Model quality 28.2/45, Memory fit 25.0/25, Speed 12.3/15, Memory use 9.8/15, workload match +5.0 | 27.0 tok/s | Runs well | 81 ptsGreat (A) · 81 of 100 points: Model quality 28.2/45, Memory fit 25.0/25, Speed 14.6/15, Memory use 9.1/15, workload match +5.0 | 38.1 tok/s | Runs well |
| Falcon-H1 34B Instruct | 80 ptsGreat (A) · 80 of 100 points: Model quality 27.4/45, Memory fit 25.0/25, Speed 10.4/15, Memory use 12.6/15, workload match +5.0 | 19.3 tok/s | Runs well | 81 ptsGreat (A) · 81 of 100 points: Model quality 27.4/45, Memory fit 25.0/25, Speed 12.4/15, Memory use 11.9/15, workload match +5.0 | 27.1 tok/s | Runs well |
| Hermes 4 14B | 80 ptsGreat (A) · 80 of 100 points: Model quality 28.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.1/15, workload match +5.0 | 43.9 tok/s | Runs well | 82 ptsGreat (A) · 82 of 100 points: Model quality 29.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.9/15, workload match +5.0 | 42.9 tok/s | Runs well |
| Ministral 3 14B | 80 ptsGreat (A) · 80 of 100 points: Model quality 27.9/45, Memory fit 25.0/25, Speed 14.7/15, Memory use 7.5/15, workload match +5.0 | 38.4 tok/s | Runs well | 81 ptsGreat (A) · 81 of 100 points: Model quality 28.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.6/15, workload match +5.0 | 45.4 tok/s | Runs well |
| Hermes 4 70B | 80 ptsGreat (A) · 80 of 100 points: Model quality 28.0/45, Memory fit 24.0/25, Speed 8.2/15, Memory use 15.0/15, workload match +5.0 | 11.9 tok/s | Tight fit | 82 ptsGreat (A) · 82 of 100 points: Model quality 28.0/45, Memory fit 25.0/25, Speed 9.7/15, Memory use 15.0/15, workload match +5.0 | 16.7 tok/s | Runs well |
| Olmo 3.1 32B Think | 80 ptsGreat (A) · 80 of 100 points: Model quality 32.5/45, Memory fit 25.0/25, Speed 10.6/15, Memory use 11.9/15 | 20.1 tok/s | Runs well | 81 ptsGreat (A) · 81 of 100 points: Model quality 32.5/45, Memory fit 25.0/25, Speed 12.6/15, Memory use 11.2/15 | 28.3 tok/s | Runs well |
| EXAONE Deep 32B | 79 ptsGreat (A) · 79 of 100 points: Model quality 31.6/45, Memory fit 25.0/25, Speed 10.7/15, Memory use 12.6/15 | 20.3 tok/s | Runs well | 81 ptsGreat (A) · 81 of 100 points: Model quality 31.6/45, Memory fit 25.0/25, Speed 12.7/15, Memory use 11.9/15 | 28.5 tok/s | Runs well |
| InternVL3 78B | 79 ptsGreat (A) · 79 of 100 points: Model quality 28.0/45, Memory fit 24.0/25, Speed 7.8/15, Memory use 15.0/15, workload match +5.0 | 10.7 tok/s | Tight fit | 81 ptsGreat (A) · 81 of 100 points: Model quality 28.0/45, Memory fit 24.0/25, Speed 9.2/15, Memory use 15.0/15, workload match +5.0 | 15.1 tok/s | Tight fit |
| InternVL3.5 8B | 79 ptsGreat (A) · 79 of 100 points: Model quality 27.9/45, Memory fit 25.0/25, Speed 14.8/15, Memory use 7.1/15, workload match +5.0 | 38.7 tok/s | Runs well | 79 ptsGreat (A) · 79 of 100 points: Model quality 27.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.4/15, workload match +5.0 | 54.5 tok/s | Runs well |
| InternVL3 14B | 79 ptsGreat (A) · 79 of 100 points: Model quality 27.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.3/15, workload match +5.0 | 42.9 tok/s | Runs well | 81 ptsGreat (A) · 81 of 100 points: Model quality 28.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.2/15, workload match +5.0 | 41.9 tok/s | Runs well |
| Qwen2.5 72B Instruct | 79 ptsGreat (A) · 79 of 100 points: Model quality 27.6/45, Memory fit 24.0/25, Speed 8.1/15, Memory use 15.0/15, workload match +5.0 | 11.5 tok/s | Tight fit | 81 ptsGreat (A) · 81 of 100 points: Model quality 27.6/45, Memory fit 24.0/25, Speed 9.6/15, Memory use 15.0/15, workload match +5.0 | 16.2 tok/s | Tight fit |
| Devstral Small 2 (24B) | 79 ptsGreat (A) · 79 of 100 points: Model quality 32.5/45, Memory fit 25.0/25, Speed 12.3/15, Memory use 9.8/15 | 27.0 tok/s | Runs well | 81 ptsGreat (A) · 81 of 100 points: Model quality 32.5/45, Memory fit 25.0/25, Speed 14.6/15, Memory use 9.1/15 | 38.0 tok/s | Runs well |
| Phi-4 (14B) | 79 ptsGreat (A) · 79 of 100 points: Model quality 27.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.3/15, workload match +5.0 | 44.2 tok/s | Runs well | 81 ptsGreat (A) · 81 of 100 points: Model quality 28.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.0/15, workload match +5.0 | 43.1 tok/s | Runs well |
| Qwen2.5-VL 72B Instruct | 79 ptsGreat (A) · 79 of 100 points: Model quality 27.6/45, Memory fit 24.0/25, Speed 8.0/15, Memory use 15.0/15, workload match +5.0 | 11.4 tok/s | Tight fit | 81 ptsGreat (A) · 81 of 100 points: Model quality 27.6/45, Memory fit 24.0/25, Speed 9.5/15, Memory use 15.0/15, workload match +5.0 | 16.1 tok/s | Tight fit |
| Mistral Small 3 24B (2501) | 79 ptsGreat (A) · 79 of 100 points: Model quality 27.4/45, Memory fit 25.0/25, Speed 12.4/15, Memory use 9.7/15, workload match +5.0 | 27.5 tok/s | Runs well | 81 ptsGreat (A) · 81 of 100 points: Model quality 27.4/45, Memory fit 25.0/25, Speed 14.8/15, Memory use 9.0/15, workload match +5.0 | 38.7 tok/s | Runs well |
| Tülu 3 70B | 79 ptsGreat (A) · 79 of 100 points: Model quality 27.2/45, Memory fit 24.0/25, Speed 8.2/15, Memory use 15.0/15, workload match +5.0 | 11.9 tok/s | Tight fit | 81 ptsGreat (A) · 81 of 100 points: Model quality 27.2/45, Memory fit 25.0/25, Speed 9.7/15, Memory use 15.0/15, workload match +5.0 | 16.7 tok/s | Runs well |
| Command R (08-2024) 32B | 79 ptsGreat (A) · 79 of 100 points: Model quality 26.5/45, Memory fit 25.0/25, Speed 10.6/15, Memory use 12.3/15, workload match +5.0 | 20.1 tok/s | Runs well | 80 ptsGreat (A) · 80 of 100 points: Model quality 26.5/45, Memory fit 25.0/25, Speed 12.6/15, Memory use 11.6/15, workload match +5.0 | 28.3 tok/s | Runs well |
| GLM-Z1 32B (0414) | 79 ptsGreat (A) · 79 of 100 points: Model quality 31.6/45, Memory fit 25.0/25, Speed 10.6/15, Memory use 11.9/15 | 19.9 tok/s | Runs well | 80 ptsGreat (A) · 80 of 100 points: Model quality 31.6/45, Memory fit 25.0/25, Speed 12.6/15, Memory use 11.2/15 | 28.0 tok/s | Runs well |
| Nous Hermes 2 Mixtral 8x7B DPO | 79 ptsGreat (A) · 79 of 100 points: Model quality 22.2/45, Memory fit 25.0/25, Speed 11.9/15, Memory use 15.0/15, workload match +5.0 | 25.2 tok/s | Runs well | 81 ptsGreat (A) · 81 of 100 points: Model quality 22.2/45, Memory fit 25.0/25, Speed 14.1/15, Memory use 15.0/15, workload match +5.0 | 35.4 tok/s | Runs well |
| DeepSeek R1 Distill Qwen 32B | 79 ptsGreat (A) · 79 of 100 points: Model quality 30.8/45, Memory fit 25.0/25, Speed 10.6/15, Memory use 12.8/15 | 19.8 tok/s | Runs well | 80 ptsGreat (A) · 80 of 100 points: Model quality 30.8/45, Memory fit 25.0/25, Speed 12.5/15, Memory use 12.1/15 | 27.8 tok/s | Runs well |
| Gemma 2 27B IT | 79 ptsGreat (A) · 79 of 100 points: Model quality 26.5/45, Memory fit 25.0/25, Speed 11.6/15, Memory use 11.0/15, workload match +5.0 | 23.9 tok/s | Runs well | 80 ptsGreat (A) · 80 of 100 points: Model quality 26.5/45, Memory fit 25.0/25, Speed 13.7/15, Memory use 10.3/15, workload match +5.0 | 33.6 tok/s | Runs well |
| Qwen3 8B | 78 ptsGreat (A) · 78 of 100 points: Model quality 27.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.9/15, workload match +5.0 | 40.3 tok/s | Runs well | 78 ptsGreat (A) · 78 of 100 points: Model quality 27.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15, workload match +5.0 | 56.8 tok/s | Runs well |
| Qwen3-VL 8B Instruct | 78 ptsGreat (A) · 78 of 100 points: Model quality 27.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15, workload match +5.0 | 51.3 tok/s | Runs well | 79 ptsGreat (A) · 79 of 100 points: Model quality 27.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.5/15, workload match +5.0 | 53.0 tok/s | Runs well |
| LFM2.5 8B-A1B | 78 ptsGreat (A) · 78 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 10.3/15, workload match +5.0 | 50.8 tok/s | Runs well | 78 ptsGreat (A) · 78 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 9.6/15, workload match +5.0 | 71.4 tok/s | Runs well |
| Qwen2.5 14B Instruct | 78 ptsGreat (A) · 78 of 100 points: Model quality 26.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.2/15, workload match +5.0 | 44.2 tok/s | Runs well | 80 ptsGreat (A) · 80 of 100 points: Model quality 27.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.0/15, workload match +5.0 | 43.1 tok/s | Runs well |
| Jamba 1.5 Mini | 78 ptsGreat (A) · 78 of 100 points: Model quality 21.4/45, Memory fit 25.0/25, Speed 12.3/15, Memory use 15.0/15, workload match +5.0 | 27.0 tok/s | Runs well | 81 ptsGreat (A) · 81 of 100 points: Model quality 21.4/45, Memory fit 25.0/25, Speed 14.6/15, Memory use 15.0/15, workload match +5.0 | 38.1 tok/s | Runs well |
| Voxtral Small 24B | 78 ptsGreat (A) · 78 of 100 points: Model quality 26.5/45, Memory fit 25.0/25, Speed 12.3/15, Memory use 9.9/15, workload match +5.0 | 26.8 tok/s | Runs well | 80 ptsGreat (A) · 80 of 100 points: Model quality 26.5/45, Memory fit 25.0/25, Speed 14.6/15, Memory use 9.2/15, workload match +5.0 | 37.6 tok/s | Runs well |
| Dolphin Mistral 24B Venice Edition | 78 ptsGreat (A) · 78 of 100 points: Model quality 26.5/45, Memory fit 25.0/25, Speed 12.3/15, Memory use 9.8/15, workload match +5.0 | 27.0 tok/s | Runs well | 80 ptsGreat (A) · 80 of 100 points: Model quality 26.5/45, Memory fit 25.0/25, Speed 14.6/15, Memory use 9.1/15, workload match +5.0 | 38.0 tok/s | Runs well |
| Aya Vision 32B | 78 ptsGreat (A) · 78 of 100 points: Model quality 25.7/45, Memory fit 25.0/25, Speed 10.5/15, Memory use 12.5/15, workload match +5.0 | 19.6 tok/s | Runs well | 79 ptsGreat (A) · 79 of 100 points: Model quality 25.7/45, Memory fit 25.0/25, Speed 12.5/15, Memory use 11.8/15, workload match +5.0 | 27.6 tok/s | Runs well |
| Llama 3.1 70B Instruct | 78 ptsGreat (A) · 78 of 100 points: Model quality 26.5/45, Memory fit 24.0/25, Speed 8.2/15, Memory use 15.0/15, workload match +5.0 | 11.9 tok/s | Tight fit | 81 ptsGreat (A) · 81 of 100 points: Model quality 26.5/45, Memory fit 25.0/25, Speed 9.7/15, Memory use 15.0/15, workload match +5.0 | 16.7 tok/s | Runs well |
| Ring Flash 2.0 (100B-A6B) | 78 ptsGreat (A) · 78 of 100 points: Model quality 24.6/45, Memory fit 24.0/25, Speed 15.0/15, Memory use 15.0/15 | 87.4 tok/s | Tight fit | 78 ptsGreat (A) · 78 of 100 points: Model quality 24.6/45, Memory fit 24.0/25, Speed 15.0/15, Memory use 15.0/15 | 123 tok/s | Tight fit |
| Aya Expanse 32B | 78 ptsGreat (A) · 78 of 100 points: Model quality 25.7/45, Memory fit 25.0/25, Speed 10.6/15, Memory use 12.3/15, workload match +5.0 | 20.1 tok/s | Runs well | 79 ptsGreat (A) · 79 of 100 points: Model quality 25.7/45, Memory fit 25.0/25, Speed 12.6/15, Memory use 11.6/15, workload match +5.0 | 28.3 tok/s | Runs well |
| OLMo 2 32B Instruct | 78 ptsGreat (A) · 78 of 100 points: Model quality 25.7/45, Memory fit 25.0/25, Speed 10.6/15, Memory use 12.1/15, workload match +5.0 | 20.1 tok/s | Runs well | 79 ptsGreat (A) · 79 of 100 points: Model quality 25.7/45, Memory fit 25.0/25, Speed 12.6/15, Memory use 11.4/15, workload match +5.0 | 28.3 tok/s | Runs well |
| Command R 35B (v01) | 78 ptsGreat (A) · 78 of 100 points: Model quality 23.1/45, Memory fit 25.0/25, Speed 10.2/15, Memory use 15.0/15, workload match +5.0 | 18.6 tok/s | Runs well | 80 ptsGreat (A) · 80 of 100 points: Model quality 23.1/45, Memory fit 25.0/25, Speed 12.1/15, Memory use 15.0/15, workload match +5.0 | 26.1 tok/s | Runs well |
| Mixtral 8x7B Instruct | 78 ptsGreat (A) · 78 of 100 points: Model quality 21.4/45, Memory fit 25.0/25, Speed 11.9/15, Memory use 15.0/15, workload match +5.0 | 25.2 tok/s | Runs well | 80 ptsGreat (A) · 80 of 100 points: Model quality 21.4/45, Memory fit 25.0/25, Speed 14.1/15, Memory use 15.0/15, workload match +5.0 | 35.4 tok/s | Runs well |
| OpenHands LM 32B | 78 ptsGreat (A) · 78 of 100 points: Model quality 29.9/45, Memory fit 25.0/25, Speed 10.6/15, Memory use 12.8/15 | 19.8 tok/s | Runs well | 79 ptsGreat (A) · 79 of 100 points: Model quality 29.9/45, Memory fit 25.0/25, Speed 12.5/15, Memory use 12.1/15 | 27.9 tok/s | Runs well |
| Kimi-VL A3B Instruct | 78 ptsGreat (A) · 78 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 10.6/15, workload match +5.0 | 55.1 tok/s | Runs well | 77 ptsGreat (A) · 77 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 9.9/15, workload match +5.0 | 77.5 tok/s | Runs well |
| GLM-4.1V 9B Thinking | 78 ptsGreat (A) · 78 of 100 points: Model quality 26.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.4/15, workload match +5.0 | 43.8 tok/s | Runs well | 78 ptsGreat (A) · 78 of 100 points: Model quality 27.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.8/15, workload match +5.0 | 45.2 tok/s | Runs well |
| Phi-4 Reasoning Plus (14B) | 78 ptsGreat (A) · 78 of 100 points: Model quality 30.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.2/15 | 44.3 tok/s | Runs well | 80 ptsGreat (A) · 80 of 100 points: Model quality 32.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.0/15 | 43.2 tok/s | Runs well |
| MiMo-VL 7B RL | 78 ptsGreat (A) · 78 of 100 points: Model quality 26.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.0/15, workload match +5.0 | 39.8 tok/s | Runs well | 77 ptsGreat (A) · 77 of 100 points: Model quality 26.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15, workload match +5.0 | 55.9 tok/s | Runs well |
| Granite 4.1 8B | 77 ptsGreat (A) · 77 of 100 points: Model quality 26.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15, workload match +5.0 | 51.2 tok/s | Runs well | 78 ptsGreat (A) · 78 of 100 points: Model quality 27.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15, workload match +5.0 | 52.9 tok/s | Runs well |
| Magistral Small 1.2 (24B) | 77 ptsGreat (A) · 77 of 100 points: Model quality 30.8/45, Memory fit 25.0/25, Speed 12.3/15, Memory use 9.8/15 | 27.0 tok/s | Runs well | 79 ptsGreat (A) · 79 of 100 points: Model quality 30.8/45, Memory fit 25.0/25, Speed 14.6/15, Memory use 9.1/15 | 38.0 tok/s | Runs well |
| Qwen2.5 14B Instruct 1M | 77 ptsGreat (A) · 77 of 100 points: Model quality 25.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.2/15, workload match +5.0 | 43.9 tok/s | Runs well | 79 ptsGreat (A) · 79 of 100 points: Model quality 26.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.0/15, workload match +5.0 | 42.9 tok/s | Runs well |
| Hermes 3 (Llama 3.1 70B) | 77 ptsGreat (A) · 77 of 100 points: Model quality 25.7/45, Memory fit 24.0/25, Speed 8.2/15, Memory use 15.0/15, workload match +5.0 | 11.9 tok/s | Tight fit | 80 ptsGreat (A) · 80 of 100 points: Model quality 25.7/45, Memory fit 25.0/25, Speed 9.7/15, Memory use 15.0/15, workload match +5.0 | 16.7 tok/s | Runs well |
| MiniCPM-V 4.5 8B | 77 ptsGreat (A) · 77 of 100 points: Model quality 26.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 51.8 tok/s | Runs well | 78 ptsGreat (A) · 78 of 100 points: Model quality 27.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.5/15, workload match +5.0 | 53.4 tok/s | Runs well |
| Gemma 3 12B IT | 77 ptsGreat (A) · 77 of 100 points: Model quality 26.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15, workload match +5.0 | 43.9 tok/s | Runs well | 78 ptsGreat (A) · 78 of 100 points: Model quality 26.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15, workload match +5.0 | 51.9 tok/s | Runs well |
| Qwen2 57B-A14B Instruct | 77 ptsGreat (A) · 77 of 100 points: Model quality 21.4/45, Memory fit 25.0/25, Speed 11.4/15, Memory use 15.0/15, workload match +5.0 | 23.2 tok/s | Runs well | 79 ptsGreat (A) · 79 of 100 points: Model quality 21.4/45, Memory fit 25.0/25, Speed 13.5/15, Memory use 15.0/15, workload match +5.0 | 32.6 tok/s | Runs well |
| Dolphin 3.0 R1 Mistral 24B | 77 ptsGreat (A) · 77 of 100 points: Model quality 25.7/45, Memory fit 25.0/25, Speed 12.4/15, Memory use 9.7/15, workload match +5.0 | 27.5 tok/s | Runs well | 79 ptsGreat (A) · 79 of 100 points: Model quality 25.7/45, Memory fit 25.0/25, Speed 14.8/15, Memory use 9.0/15, workload match +5.0 | 38.7 tok/s | Runs well |
| Solar Pro Preview 22B | 77 ptsGreat (A) · 77 of 100 points: Model quality 25.7/45, Memory fit 25.0/25, Speed 12.8/15, Memory use 9.3/15, workload match +5.0 | 29.3 tok/s | Runs well | 79 ptsGreat (A) · 79 of 100 points: Model quality 25.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.6/15, workload match +5.0 | 41.3 tok/s | Runs well |
| QVQ 72B Preview | 77 ptsGreat (A) · 77 of 100 points: Model quality 25.7/45, Memory fit 24.0/25, Speed 8.0/15, Memory use 15.0/15, workload match +5.0 | 11.4 tok/s | Tight fit | 79 ptsGreat (A) · 79 of 100 points: Model quality 25.7/45, Memory fit 24.0/25, Speed 9.5/15, Memory use 15.0/15, workload match +5.0 | 16.1 tok/s | Tight fit |
| Gemma 4 E2B | 77 ptsGreat (A) · 77 of 100 points: Model quality 27.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.8/15, workload match +5.0 | 64.5 tok/s | Runs well | 79 ptsGreat (A) · 79 of 100 points: Model quality 27.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.4/15, workload match +5.0 | 41.9 tok/s | Runs well |
| Qwen3.5 4B | 77 ptsGreat (A) · 77 of 100 points: Model quality 27.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.6/15, workload match +5.0 | 70.9 tok/s | Runs well | 78 ptsGreat (A) · 78 of 100 points: Model quality 27.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 46.0 tok/s | Runs well |
| Nemotron Nano 9B v2 | 77 ptsGreat (A) · 77 of 100 points: Model quality 26.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.7/15, workload match +5.0 | 50.7 tok/s | Runs well | 78 ptsGreat (A) · 78 of 100 points: Model quality 27.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.0/15, workload match +5.0 | 52.3 tok/s | Runs well |
| Kimi Dev 72B | 77 ptsGreat (A) · 77 of 100 points: Model quality 30.2/45, Memory fit 24.0/25, Speed 8.1/15, Memory use 15.0/15 | 11.5 tok/s | Tight fit | 78 ptsGreat (A) · 78 of 100 points: Model quality 30.2/45, Memory fit 24.0/25, Speed 9.6/15, Memory use 15.0/15 | 16.2 tok/s | Tight fit |
| Phi-4 Reasoning (14B) | 77 ptsGreat (A) · 77 of 100 points: Model quality 29.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.2/15 | 44.3 tok/s | Runs well | 79 ptsGreat (A) · 79 of 100 points: Model quality 31.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.0/15 | 43.2 tok/s | Runs well |
| UI-Mate 9B | 77 ptsGreat (A) · 77 of 100 points: Model quality 31.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.0/15 | 47.9 tok/s | Runs well | 77 ptsGreat (A) · 77 of 100 points: Model quality 31.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.4/15 | 49.4 tok/s | Runs well |
| UI-Venus 2 9B | 77 ptsGreat (A) · 77 of 100 points: Model quality 31.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.0/15 | 47.9 tok/s | Runs well | 77 ptsGreat (A) · 77 of 100 points: Model quality 31.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.4/15 | 49.4 tok/s | Runs well |
| Llama 3.1 Nemotron 70B Instruct | 77 ptsGreat (A) · 77 of 100 points: Model quality 24.9/45, Memory fit 24.0/25, Speed 8.2/15, Memory use 15.0/15, workload match +5.0 | 11.9 tok/s | Tight fit | 79 ptsGreat (A) · 79 of 100 points: Model quality 24.9/45, Memory fit 25.0/25, Speed 9.7/15, Memory use 15.0/15, workload match +5.0 | 16.7 tok/s | Runs well |
| Devstral Small 1.1 (24B) | 77 ptsGreat (A) · 77 of 100 points: Model quality 29.9/45, Memory fit 25.0/25, Speed 12.4/15, Memory use 9.7/15 | 27.5 tok/s | Runs well | 78 ptsGreat (A) · 78 of 100 points: Model quality 29.9/45, Memory fit 25.0/25, Speed 14.8/15, Memory use 9.0/15 | 38.7 tok/s | Runs well |
| Apertus v1.5 70B | 77 ptsGreat (A) · 77 of 100 points: Model quality 24.9/45, Memory fit 24.0/25, Speed 8.1/15, Memory use 15.0/15, workload match +5.0 | 11.7 tok/s | Tight fit | 78 ptsGreat (A) · 78 of 100 points: Model quality 24.9/45, Memory fit 24.0/25, Speed 9.6/15, Memory use 15.0/15, workload match +5.0 | 16.4 tok/s | Tight fit |
| Qwen2.5 Coder 32B Instruct | 76 ptsGreat (A) · 76 of 100 points: Model quality 28.6/45, Memory fit 25.0/25, Speed 10.6/15, Memory use 12.8/15 | 19.8 tok/s | Runs well | 78 ptsGreat (A) · 78 of 100 points: Model quality 28.6/45, Memory fit 25.0/25, Speed 12.5/15, Memory use 12.1/15 | 27.8 tok/s | Runs well |
| TMax 4B | 76 ptsGreat (A) · 76 of 100 points: Model quality 27.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.4/15, workload match +5.0 | 78.5 tok/s | Runs well | 78 ptsGreat (A) · 78 of 100 points: Model quality 27.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15, workload match +5.0 | 50.9 tok/s | Runs well |
| InternVL3 8B | 76 ptsGreat (A) · 76 of 100 points: Model quality 25.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.4/15, workload match +5.0 | 41.6 tok/s | Runs well | 75 ptsGreat (A) · 75 of 100 points: Model quality 25.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.7/15, workload match +5.0 | 58.6 tok/s | Runs well |
| Aya 23 35B | 76 ptsGreat (A) · 76 of 100 points: Model quality 21.4/45, Memory fit 25.0/25, Speed 10.2/15, Memory use 15.0/15, workload match +5.0 | 18.6 tok/s | Runs well | 78 ptsGreat (A) · 78 of 100 points: Model quality 21.4/45, Memory fit 25.0/25, Speed 12.1/15, Memory use 15.0/15, workload match +5.0 | 26.1 tok/s | Runs well |
| Agents-A1 4B | 76 ptsGreat (A) · 76 of 100 points: Model quality 27.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.6/15, workload match +5.0 | 72.8 tok/s | Runs well | 77 ptsGreat (A) · 77 of 100 points: Model quality 27.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15, workload match +5.0 | 47.2 tok/s | Runs well |
| QwQ 32B Preview | 76 ptsGreat (A) · 76 of 100 points: Model quality 28.2/45, Memory fit 25.0/25, Speed 10.6/15, Memory use 12.8/15 | 19.8 tok/s | Runs well | 77 ptsGreat (A) · 77 of 100 points: Model quality 28.2/45, Memory fit 25.0/25, Speed 12.5/15, Memory use 12.1/15 | 27.9 tok/s | Runs well |
| Qwen2 72B Instruct | 76 ptsGreat (A) · 76 of 100 points: Model quality 24.2/45, Memory fit 24.0/25, Speed 8.1/15, Memory use 15.0/15, workload match +5.0 | 11.5 tok/s | Tight fit | 77 ptsGreat (A) · 77 of 100 points: Model quality 24.2/45, Memory fit 24.0/25, Speed 9.6/15, Memory use 15.0/15, workload match +5.0 | 16.2 tok/s | Tight fit |
| Qwen2-VL 72B Instruct | 76 ptsGreat (A) · 76 of 100 points: Model quality 24.2/45, Memory fit 24.0/25, Speed 8.0/15, Memory use 15.0/15, workload match +5.0 | 11.4 tok/s | Tight fit | 77 ptsGreat (A) · 77 of 100 points: Model quality 24.2/45, Memory fit 24.0/25, Speed 9.5/15, Memory use 15.0/15, workload match +5.0 | 16.1 tok/s | Tight fit |
| Ministral 3 8B | 76 ptsGreat (A) · 76 of 100 points: Model quality 24.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15, workload match +5.0 | 50.5 tok/s | Runs well | 76 ptsGreat (A) · 76 of 100 points: Model quality 25.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.5/15, workload match +5.0 | 52.1 tok/s | Runs well |
| Moonlight 16B-A3B Instruct | 76 ptsGreat (A) · 76 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 10.4/15, workload match +5.0 | 55.1 tok/s | Runs well | 75 ptsGreat (A) · 75 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 9.7/15, workload match +5.0 | 77.5 tok/s | Runs well |
| Ministral 3 14B Reasoning | 76 ptsGreat (A) · 76 of 100 points: Model quality 28.8/45, Memory fit 25.0/25, Speed 14.7/15, Memory use 7.5/15 | 38.4 tok/s | Runs well | 76 ptsGreat (A) · 76 of 100 points: Model quality 29.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.6/15 | 45.4 tok/s | Runs well |
| Llama 3.1 Nemotron Nano 8B v1 | 76 ptsGreat (A) · 76 of 100 points: Model quality 24.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.7/15, workload match +5.0 | 41.1 tok/s | Runs well | 75 ptsGreat (A) · 75 of 100 points: Model quality 24.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 57.9 tok/s | Runs well |
| MiniCPM4 8B | 75 ptsGreat (A) · 75 of 100 points: Model quality 24.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.4/15, workload match +5.0 | 40.3 tok/s | Runs well | 75 ptsGreat (A) · 75 of 100 points: Model quality 24.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.7/15, workload match +5.0 | 56.8 tok/s | Runs well |
| Fara 1.5 4B | 75 ptsGreat (A) · 75 of 100 points: Model quality 26.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.6/15, workload match +5.0 | 72.8 tok/s | Runs well | 77 ptsGreat (A) · 77 of 100 points: Model quality 26.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15, workload match +5.0 | 47.2 tok/s | Runs well |
| Apriel Nemotron 15B Thinker | 75 ptsGreat (A) · 75 of 100 points: Model quality 28.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.3/15 | 43.4 tok/s | Runs well | 77 ptsGreat (A) · 77 of 100 points: Model quality 29.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.2/15 | 42.3 tok/s | Runs well |
| Mistral Small 22B (2409) | 75 ptsGreat (A) · 75 of 100 points: Model quality 23.1/45, Memory fit 25.0/25, Speed 12.8/15, Memory use 9.6/15, workload match +5.0 | 29.2 tok/s | Runs well | 76 ptsGreat (A) · 76 of 100 points: Model quality 23.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.9/15, workload match +5.0 | 41.0 tok/s | Runs well |
| AceReason Nemotron 14B | 75 ptsGreat (A) · 75 of 100 points: Model quality 28.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.2/15 | 43.9 tok/s | Runs well | 77 ptsGreat (A) · 77 of 100 points: Model quality 29.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.0/15 | 42.9 tok/s | Runs well |
| OpenReasoning Nemotron 14B | 75 ptsGreat (A) · 75 of 100 points: Model quality 28.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.2/15 | 43.9 tok/s | Runs well | 77 ptsGreat (A) · 77 of 100 points: Model quality 29.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.0/15 | 42.9 tok/s | Runs well |
| Qwen2.5 Coder 14B Instruct | 75 ptsGreat (A) · 75 of 100 points: Model quality 28.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.2/15 | 43.9 tok/s | Runs well | 77 ptsGreat (A) · 77 of 100 points: Model quality 29.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.0/15 | 42.9 tok/s | Runs well |
| Devstral Small 2505 | 75 ptsGreat (A) · 75 of 100 points: Model quality 28.2/45, Memory fit 25.0/25, Speed 12.4/15, Memory use 9.7/15 | 27.5 tok/s | Runs well | 76 ptsGreat (A) · 76 of 100 points: Model quality 28.2/45, Memory fit 25.0/25, Speed 14.8/15, Memory use 9.0/15 | 38.7 tok/s | Runs well |
| Mage-VL 4.7B | 75 ptsGreat (A) · 75 of 100 points: Model quality 25.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.1/15, workload match +5.0 | 69.7 tok/s | Runs well | 76 ptsGreat (A) · 76 of 100 points: Model quality 25.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15, workload match +5.0 | 45.2 tok/s | Runs well |
| Reka Flash 3.1 (21B) | 75 ptsGreat (A) · 75 of 100 points: Model quality 28.2/45, Memory fit 25.0/25, Speed 13.2/15, Memory use 8.8/15 | 31.0 tok/s | Runs well | 76 ptsGreat (A) · 76 of 100 points: Model quality 28.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.1/15 | 43.7 tok/s | Runs well |
| Falcon 3 10B Instruct | 75 ptsGreat (A) · 75 of 100 points: Model quality 23.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.8/15, workload match +5.0 | 43.7 tok/s | Runs well | 75 ptsGreat (A) · 75 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.3/15, workload match +5.0 | 45.1 tok/s | Runs well |
| Qwen2.5-VL 7B Instruct | 74 ptsGreat (A) · 74 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15, workload match +5.0 | 39.9 tok/s | Runs well | 74 ptsGreat (A) · 74 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15, workload match +5.0 | 56.1 tok/s | Runs well |
| Hunyuan 7B Instruct | 74 ptsGreat (A) · 74 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.5/15, workload match +5.0 | 44.1 tok/s | Runs well | 74 ptsGreat (A) · 74 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.8/15, workload match +5.0 | 62.0 tok/s | Runs well |
| Apertus 70B Instruct | 74 ptsGreat (A) · 74 of 100 points: Model quality 22.7/45, Memory fit 24.0/25, Speed 8.2/15, Memory use 15.0/15, workload match +5.0 | 11.9 tok/s | Tight fit | 77 ptsGreat (A) · 77 of 100 points: Model quality 22.7/45, Memory fit 25.0/25, Speed 9.7/15, Memory use 15.0/15, workload match +5.0 | 16.7 tok/s | Runs well |
| DeepSeek R1 0528 Qwen3 8B | 74 ptsGreat (A) · 74 of 100 points: Model quality 27.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.9/15 | 40.3 tok/s | Runs well | 74 ptsGreat (A) · 74 of 100 points: Model quality 27.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15 | 56.8 tok/s | Runs well |
| Qwen2.5 7B Instruct | 74 ptsGreat (A) · 74 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15, workload match +5.0 | 43.4 tok/s | Runs well | 73 ptsGreat (A) · 73 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15, workload match +5.0 | 61.0 tok/s | Runs well |
| Mistral Nemo 12B Instruct | 74 ptsGreat (A) · 74 of 100 points: Model quality 22.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.9/15, workload match +5.0 | 43.9 tok/s | Runs well | 75 ptsGreat (A) · 75 of 100 points: Model quality 23.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.9/15, workload match +5.0 | 51.9 tok/s | Runs well |
| DeepCoder 14B Preview | 74 ptsGreat (A) · 74 of 100 points: Model quality 27.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.2/15 | 43.9 tok/s | Runs well | 76 ptsGreat (A) · 76 of 100 points: Model quality 28.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.0/15 | 42.9 tok/s | Runs well |
| DeepSeek R1 Distill Llama 70B | 74 ptsGreat (A) · 74 of 100 points: Model quality 27.2/45, Memory fit 24.0/25, Speed 8.2/15, Memory use 15.0/15 | 11.9 tok/s | Tight fit | 76 ptsGreat (A) · 76 of 100 points: Model quality 27.2/45, Memory fit 25.0/25, Speed 9.7/15, Memory use 15.0/15 | 16.7 tok/s | Runs well |
| Yi 1.5 34B Chat | 74 ptsGreat (A) · 74 of 100 points: Model quality 21.4/45, Memory fit 25.0/25, Speed 10.3/15, Memory use 12.7/15, workload match +5.0 | 18.9 tok/s | Runs well | 75 ptsGreat (A) · 75 of 100 points: Model quality 21.4/45, Memory fit 25.0/25, Speed 12.2/15, Memory use 12.0/15, workload match +5.0 | 26.6 tok/s | Runs well |
| Granite 4.2 3B | 74 ptsGreat (A) · 74 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.0/15, workload match +5.0 | 41.6 tok/s | Runs well | 73 ptsGreat (A) · 73 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.3/15, workload match +5.0 | 58.5 tok/s | Runs well |
| Qwen3-VL 4B Instruct | 74 ptsGreat (A) · 74 of 100 points: Model quality 24.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.0/15, workload match +5.0 | 74.4 tok/s | Runs well | 75 ptsGreat (A) · 75 of 100 points: Model quality 24.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15, workload match +5.0 | 48.3 tok/s | Runs well |
| InternVL2.5 8B | 74 ptsGreat (A) · 74 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.8/15, workload match +5.0 | 40.9 tok/s | Runs well | 73 ptsGreat (A) · 73 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 57.5 tok/s | Runs well |
| GLM-4 9B 0414 | 74 ptsGreat (A) · 74 of 100 points: Model quality 23.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.0/15, workload match +5.0 | 47.9 tok/s | Runs well | 74 ptsGreat (A) · 74 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.4/15, workload match +5.0 | 49.5 tok/s | Runs well |
| Llama 3 70B Instruct | 74 ptsGreat (A) · 74 of 100 points: Model quality 21.9/45, Memory fit 24.0/25, Speed 8.2/15, Memory use 15.0/15, workload match +5.0 | 11.9 tok/s | Tight fit | 76 ptsGreat (A) · 76 of 100 points: Model quality 21.9/45, Memory fit 25.0/25, Speed 9.7/15, Memory use 15.0/15, workload match +5.0 | 16.7 tok/s | Runs well |
| Olmo 3 7B Instruct | 74 ptsGreat (A) · 74 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15, workload match +5.0 | 45.3 tok/s | Runs well | 73 ptsGreat (A) · 73 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15, workload match +5.0 | 63.7 tok/s | Runs well |
| Qwen3 4B Instruct 2507 | 74 ptsGreat (A) · 74 of 100 points: Model quality 24.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.8/15, workload match +5.0 | 82.2 tok/s | Runs well | 75 ptsGreat (A) · 75 of 100 points: Model quality 24.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15, workload match +5.0 | 53.3 tok/s | Runs well |
| MiniCPM-o 2.6 8B | 73 ptsGreat (A) · 73 of 100 points: Model quality 23.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.8/15, workload match +5.0 | 51.9 tok/s | Runs well | 74 ptsGreat (A) · 74 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 53.6 tok/s | Runs well |
| Codestral 22B | 73 ptsGreat (A) · 73 of 100 points: Model quality 26.5/45, Memory fit 25.0/25, Speed 12.8/15, Memory use 9.6/15 | 29.2 tok/s | Runs well | 75 ptsGreat (A) · 75 of 100 points: Model quality 26.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.9/15 | 41.0 tok/s | Runs well |
| Pixtral 12B (2409) | 73 ptsGreat (A) · 73 of 100 points: Model quality 21.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.0/15, workload match +5.0 | 43.2 tok/s | Runs well | 74 ptsGreat (A) · 74 of 100 points: Model quality 22.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.0/15, workload match +5.0 | 51.1 tok/s | Runs well |
| Gemma 2 9B IT | 73 ptsGreat (A) · 73 of 100 points: Model quality 22.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15, workload match +5.0 | 48.7 tok/s | Runs well | 74 ptsGreat (A) · 74 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.9/15, workload match +5.0 | 50.3 tok/s | Runs well |
| DeepSeek R1 Distill Qwen 14B | 73 ptsGreat (A) · 73 of 100 points: Model quality 26.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.2/15 | 43.9 tok/s | Runs well | 75 ptsGreat (A) · 75 of 100 points: Model quality 27.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.0/15 | 42.9 tok/s | Runs well |
| Qwen2.5 7B Instruct 1M | 73 ptsGreat (A) · 73 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15, workload match +5.0 | 43.4 tok/s | Runs well | 73 ptsGreat (A) · 73 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15, workload match +5.0 | 61.0 tok/s | Runs well |
| Falcon-H1 7B Instruct | 73 ptsGreat (A) · 73 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15, workload match +5.0 | 43.5 tok/s | Runs well | 72 ptsGreat (A) · 72 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.5/15, workload match +5.0 | 61.3 tok/s | Runs well |
| InternLM2.5 20B Chat | 73 ptsGreat (A) · 73 of 100 points: Model quality 21.4/45, Memory fit 25.0/25, Speed 13.6/15, Memory use 8.7/15, workload match +5.0 | 32.7 tok/s | Runs well | 75 ptsGreat (A) · 75 of 100 points: Model quality 21.8/45, Memory fit 25.0/25, Speed 14.6/15, Memory use 9.0/15, workload match +5.0 | 38.0 tok/s | Runs well |
| DeepSeek Coder V2 Lite (16B) | 73 ptsGreat (A) · 73 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 10.3/15 | 68.8 tok/s | Runs well | 78 ptsGreat (A) · 78 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | 44.6 tok/s | Runs well |
| Granite 4.0 H Tiny (7B-A1B) | 73 ptsGreat (A) · 73 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.8/15, workload match +5.0 | 76.1 tok/s | Runs well | 72 ptsGreat (A) · 72 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.1/15, workload match +5.0 | 107 tok/s | Runs well |
| JanusCoder 14B | 73 ptsGreat (A) · 73 of 100 points: Model quality 26.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.1/15 | 43.9 tok/s | Runs well | 75 ptsGreat (A) · 75 of 100 points: Model quality 27.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.9/15 | 42.9 tok/s | Runs well |
| Hermes 3 (Llama 3.1 8B) | 73 ptsGreat (A) · 73 of 100 points: Model quality 21.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.7/15, workload match +5.0 | 41.1 tok/s | Runs well | 72 ptsGreat (A) · 72 of 100 points: Model quality 21.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 57.9 tok/s | Runs well |
| Llama 3.1 8B Instruct | 73 ptsGreat (A) · 73 of 100 points: Model quality 21.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.7/15, workload match +5.0 | 41.1 tok/s | Runs well | 72 ptsGreat (A) · 72 of 100 points: Model quality 21.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 57.9 tok/s | Runs well |
| DeepSeek V2 Lite Chat | 73 ptsGreat (A) · 73 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 10.3/15, workload match +5.0 | 68.8 tok/s | Runs well | 78 ptsGreat (A) · 78 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 44.6 tok/s | Runs well |
| MiniCPM-V 2.6 8B | 73 ptsGreat (A) · 73 of 100 points: Model quality 21.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.5/15, workload match +5.0 | 40.8 tok/s | Runs well | 72 ptsGreat (A) · 72 of 100 points: Model quality 21.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.8/15, workload match +5.0 | 57.4 tok/s | Runs well |
| InternLM3 8B Instruct | 73 ptsGreat (A) · 73 of 100 points: Model quality 22.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.8/15, workload match +5.0 | 51.2 tok/s | Runs well | 73 ptsGreat (A) · 73 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 52.8 tok/s | Runs well |
| Llama 3.2 90B Vision Instruct | 73 ptsGreat (A) · 73 of 100 points: Model quality 20.8/45, Memory fit 24.0/25, Speed 8.2/15, Memory use 15.0/15, workload match +5.0 | 12.0 tok/s | Tight fit | 74 ptsGreat (A) · 74 of 100 points: Model quality 20.8/45, Memory fit 24.0/25, Speed 9.8/15, Memory use 15.0/15, workload match +5.0 | 16.9 tok/s | Tight fit |
| Seed-Coder 8B Reasoning | 72 ptsGreat (A) · 72 of 100 points: Model quality 26.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.9/15 | 40.0 tok/s | Runs well | 72 ptsGreat (A) · 72 of 100 points: Model quality 26.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15 | 56.4 tok/s | Runs well |
| Granite 3.2 8B Instruct | 72 ptsGreat (A) · 72 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.9/15, workload match +5.0 | 40.4 tok/s | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15, workload match +5.0 | 56.9 tok/s | Runs well |
| Granite 3.3 8B | 72 ptsGreat (A) · 72 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.9/15, workload match +5.0 | 40.4 tok/s | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15, workload match +5.0 | 56.9 tok/s | Runs well |
| GLM-4 9B Chat 1M | 72 ptsGreat (A) · 72 of 100 points: Model quality 21.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15, workload match +5.0 | 47.5 tok/s | Runs well | 73 ptsGreat (A) · 73 of 100 points: Model quality 21.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15, workload match +5.0 | 49.0 tok/s | Runs well |
| Ministral 8B (2410) | 72 ptsGreat (A) · 72 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.8/15, workload match +5.0 | 41.2 tok/s | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 58.0 tok/s | Runs well |
| Phi-3 Medium 128K Instruct | 72 ptsGreat (A) · 72 of 100 points: Model quality 20.1/45, Memory fit 25.0/25, Speed 14.7/15, Memory use 7.7/15, workload match +5.0 | 38.4 tok/s | Runs well | 73 ptsGreat (A) · 73 of 100 points: Model quality 20.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.8/15, workload match +5.0 | 45.4 tok/s | Runs well |
| OLMo 2 13B Instruct | 72 ptsGreat (A) · 72 of 100 points: Model quality 19.2/45, Memory fit 25.0/25, Speed 14.8/15, Memory use 8.4/15, workload match +5.0 | 39.1 tok/s | Runs well | 73 ptsGreat (A) · 73 of 100 points: Model quality 19.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.5/15, workload match +5.0 | 46.2 tok/s | Runs well |
| DeepHermes 3 Llama 3 8B Preview | 72 ptsGreat (A) · 72 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.7/15, workload match +5.0 | 41.1 tok/s | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 57.9 tok/s | Runs well |
| Dolphin 3.0 Llama 3.1 8B | 72 ptsGreat (A) · 72 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.7/15, workload match +5.0 | 41.1 tok/s | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 57.9 tok/s | Runs well |
| Tülu 3 8B | 72 ptsGreat (A) · 72 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.7/15, workload match +5.0 | 41.1 tok/s | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 57.9 tok/s | Runs well |
| GLM-4 9B Chat (2024) | 72 ptsGreat (A) · 72 of 100 points: Model quality 21.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.0/15, workload match +5.0 | 47.9 tok/s | Runs well | 72 ptsGreat (A) · 72 of 100 points: Model quality 21.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.4/15, workload match +5.0 | 49.5 tok/s | Runs well |
| EXAONE 3.5 7.8B | 72 ptsGreat (A) · 72 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15, workload match +5.0 | 42.3 tok/s | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15, workload match +5.0 | 59.5 tok/s | Runs well |
| LFM2 8B-A1B | 72 ptsGreat (A) · 72 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 10.2/15, workload match +5.0 | 50.8 tok/s | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 9.5/15, workload match +5.0 | 71.4 tok/s | Runs well |
| Qwen3 4B | 72 ptsGreat (A) · 72 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.8/15, workload match +5.0 | 82.2 tok/s | Runs well | 73 ptsGreat (A) · 73 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15, workload match +5.0 | 53.3 tok/s | Runs well |
| Qwen1.5 32B Chat | 72 ptsGreat (A) · 72 of 100 points: Model quality 18.8/45, Memory fit 25.0/25, Speed 10.6/15, Memory use 12.7/15, workload match +5.0 | 20.0 tok/s | Runs well | 73 ptsGreat (A) · 73 of 100 points: Model quality 18.8/45, Memory fit 25.0/25, Speed 12.6/15, Memory use 12.0/15, workload match +5.0 | 28.1 tok/s | Runs well |
| Falcon 3 7B Instruct | 72 ptsGreat (A) · 72 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.4/15, workload match +5.0 | 44.3 tok/s | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.7/15, workload match +5.0 | 62.3 tok/s | Runs well |
| Qwen2.5-Omni 7B | 72 ptsGreat (A) · 72 of 100 points: Model quality 20.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15, workload match +5.0 | 42.0 tok/s | Runs well | 72 ptsGreat (A) · 72 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.1/15, workload match +5.0 | 43.3 tok/s | Runs well |
| Phi-3 Medium (14B) | 72 ptsGreat (A) · 72 of 100 points: Model quality 20.1/45, Memory fit 25.0/25, Speed 14.7/15, Memory use 7.3/15, workload match +5.0 | 38.4 tok/s | Runs well | 72 ptsGreat (A) · 72 of 100 points: Model quality 20.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.3/15, workload match +5.0 | 45.4 tok/s | Runs well |
| Seed-Coder 8B Instruct | 72 ptsGreat (A) · 72 of 100 points: Model quality 25.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.9/15 | 40.0 tok/s | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 25.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15 | 56.4 tok/s | Runs well |
| MiMo 7B RL | 71 ptsGood (B) · 71 of 100 points: Model quality 25.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.7/15 | 42.2 tok/s | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 25.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.0/15 | 59.4 tok/s | Runs well |
| Gemma 3n E4B | 71 ptsGood (B) · 71 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15, workload match +5.0 | 42.1 tok/s | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.5/15, workload match +5.0 | 59.2 tok/s | Runs well |
| LLaVA 1.6 34B | 71 ptsGood (B) · 71 of 100 points: Model quality 18.8/45, Memory fit 25.0/25, Speed 10.2/15, Memory use 12.8/15, workload match +5.0 | 18.7 tok/s | Runs well | 73 ptsGreat (A) · 73 of 100 points: Model quality 18.8/45, Memory fit 25.0/25, Speed 12.2/15, Memory use 12.1/15, workload match +5.0 | 26.3 tok/s | Runs well |
| GLM-Z1 9B (0414) | 71 ptsGood (B) · 71 of 100 points: Model quality 25.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.0/15 | 47.9 tok/s | Runs well | 72 ptsGreat (A) · 72 of 100 points: Model quality 26.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.4/15 | 49.5 tok/s | Runs well |
| EXAONE Deep 7.8B | 71 ptsGood (B) · 71 of 100 points: Model quality 25.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15 | 42.3 tok/s | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 25.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15 | 59.5 tok/s | Runs well |
| Granite 3.1 8B Instruct | 71 ptsGood (B) · 71 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.9/15, workload match +5.0 | 40.4 tok/s | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15, workload match +5.0 | 56.9 tok/s | Runs well |
| Nemotron 3 Nano 4B | 71 ptsGood (B) · 71 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.2/15, workload match +5.0 | 83.2 tok/s | Runs well | 72 ptsGreat (A) · 72 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.3/15, workload match +5.0 | 54.0 tok/s | Runs well |
| Apertus v1.5 8B | 71 ptsGood (B) · 71 of 100 points: Model quality 20.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15, workload match +5.0 | 50.6 tok/s | Runs well | 72 ptsGreat (A) · 72 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.5/15, workload match +5.0 | 52.2 tok/s | Runs well |
| InternVL2 8B | 71 ptsGood (B) · 71 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.8/15, workload match +5.0 | 40.9 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 57.5 tok/s | Runs well |
| Dolphin 2.9.4 Llama 3.1 8B | 71 ptsGood (B) · 71 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.7/15, workload match +5.0 | 41.1 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 57.9 tok/s | Runs well |
| Hermes 2 Pro Llama 3 8B | 71 ptsGood (B) · 71 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.7/15, workload match +5.0 | 41.1 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 57.9 tok/s | Runs well |
| OpenChat 3.6 8B | 71 ptsGood (B) · 71 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.7/15, workload match +5.0 | 41.1 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 57.9 tok/s | Runs well |
| Granite 4.1 3B | 71 ptsGood (B) · 71 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.7/15, workload match +5.0 | 44.8 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.0/15, workload match +5.0 | 63.0 tok/s | Runs well |
| Falcon-H1R 7B | 71 ptsGood (B) · 71 of 100 points: Model quality 25.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15 | 43.5 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 25.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.5/15 | 61.3 tok/s | Runs well |
| Qwen2-VL 7B Instruct | 71 ptsGood (B) · 71 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15, workload match +5.0 | 39.9 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15, workload match +5.0 | 56.1 tok/s | Runs well |
| Phi-3 Small 8K Instruct | 71 ptsGood (B) · 71 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.4/15, workload match +5.0 | 44.7 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.7/15, workload match +5.0 | 62.9 tok/s | Runs well |
| Command R7B | 71 ptsGood (B) · 71 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.4/15, workload match +5.0 | 41.1 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.7/15, workload match +5.0 | 57.9 tok/s | Runs well |
| Qwen2 7B Instruct | 71 ptsGood (B) · 71 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15, workload match +5.0 | 43.4 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15, workload match +5.0 | 61.0 tok/s | Runs well |
| Qwen2.5 Coder 7B Instruct | 70 ptsGood (B) · 70 of 100 points: Model quality 24.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15 | 43.4 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 24.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15 | 61.0 tok/s | Runs well |
| Olmo 3 7B Think | 70 ptsGood (B) · 70 of 100 points: Model quality 24.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15 | 45.3 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 24.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15 | 63.7 tok/s | Runs well |
| Mistral-NeMo-Minitron 8B Instruct | 70 ptsGood (B) · 70 of 100 points: Model quality 18.9/45, Memory fit 25.0/25, Speed 14.9/15, Memory use 7.1/15, workload match +5.0 | 39.3 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 18.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.4/15, workload match +5.0 | 55.3 tok/s | Runs well |
| GLM-4.6V | 70 ptsGood (B) · 70 of 100 points: Model quality 26.1/45, Memory fit 10.0/25, Speed 14.6/15, Memory use 15.0/15, workload match +5.0 | 37.8 tok/s | Runs with offload | 85 ptsExcellent (S) · 85 of 100 points: Model quality 26.1/45, Memory fit 24.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 62.5 tok/s | Tight fit |
| Llama 3.2 11B Vision Instruct | 70 ptsGood (B) · 70 of 100 points: Model quality 18.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.0/15, workload match +5.0 | 42.2 tok/s | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 18.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.5/15, workload match +5.0 | 43.6 tok/s | Runs well |
| Llama 3 8B Instruct | 70 ptsGood (B) · 70 of 100 points: Model quality 18.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.7/15, workload match +5.0 | 41.1 tok/s | Runs well | 69 ptsGood (B) · 69 of 100 points: Model quality 18.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 57.9 tok/s | Runs well |
| OpenReasoning Nemotron 7B | 70 ptsGood (B) · 70 of 100 points: Model quality 24.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15 | 43.4 tok/s | Runs well | 69 ptsGood (B) · 69 of 100 points: Model quality 24.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15 | 61.0 tok/s | Runs well |
| Mistral Small 4 (119B-A6B) | 70 ptsGood (B) · 70 of 100 points: Model quality 25.4/45, Memory fit 10.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 75.5 tok/s | Runs with offload | 73 ptsGreat (A) · 73 of 100 points: Model quality 30.5/45, Memory fit 10.0/25, Speed 12.8/15, Memory use 15.0/15, workload match +5.0 | 29.1 tok/s | Runs with offload |
| Aya Vision 8B | 70 ptsGood (B) · 70 of 100 points: Model quality 19.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.7/15, workload match +5.0 | 52.2 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.0/15, workload match +5.0 | 53.9 tok/s | Runs well |
| Command R7B Arabic | 70 ptsGood (B) · 70 of 100 points: Model quality 18.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.4/15, workload match +5.0 | 41.1 tok/s | Runs well | 69 ptsGood (B) · 69 of 100 points: Model quality 18.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.7/15, workload match +5.0 | 57.9 tok/s | Runs well |
| Phi-4 Multimodal Instruct | 70 ptsGood (B) · 70 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.5/15, workload match +5.0 | 59.3 tok/s | Runs well | 69 ptsGood (B) · 69 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.8/15, workload match +5.0 | 83.5 tok/s | Runs well |
| InternLM2 Chat 20B | 70 ptsGood (B) · 70 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 13.6/15, Memory use 8.7/15, workload match +5.0 | 32.7 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.0/15, workload match +5.0 | 46.0 tok/s | Runs well |
| Yi 34B Chat | 70 ptsGood (B) · 70 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 10.3/15, Memory use 12.7/15, workload match +5.0 | 18.9 tok/s | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 12.2/15, Memory use 12.0/15, workload match +5.0 | 26.6 tok/s | Runs well |
| Jamba Reasoning 3B | 70 ptsGood (B) · 70 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.2/15, workload match +5.0 | 47.6 tok/s | Runs well | 69 ptsGood (B) · 69 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.5/15, workload match +5.0 | 67.0 tok/s | Runs well |
| LFM2.5-VL 3B | 69 ptsGood (B) · 69 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.2/15, workload match +5.0 | 48.8 tok/s | Runs well | 69 ptsGood (B) · 69 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.5/15, workload match +5.0 | 68.7 tok/s | Runs well |
| Qwen3 4B Thinking 2507 | 69 ptsGood (B) · 69 of 100 points: Model quality 25.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.8/15 | 82.2 tok/s | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 25.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15 | 53.3 tok/s | Runs well |
| CogVLM2 Llama 3 Chat 19B | 69 ptsGood (B) · 69 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 13.7/15, Memory use 9.9/15, workload match +5.0 | 33.3 tok/s | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 16.6/45, Memory fit 25.0/25, Speed 14.8/15, Memory use 10.2/15, workload match +5.0 | 38.7 tok/s | Runs well |
| GLM-4.5V | 69 ptsGood (B) · 69 of 100 points: Model quality 25.2/45, Memory fit 10.0/25, Speed 14.6/15, Memory use 15.0/15, workload match +5.0 | 37.8 tok/s | Runs with offload | 84 ptsGreat (A) · 84 of 100 points: Model quality 25.2/45, Memory fit 24.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 62.5 tok/s | Tight fit |
| North Micro Vision | 69 ptsGood (B) · 69 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.0/15, workload match +5.0 | 61.4 tok/s | Runs well | 69 ptsGood (B) · 69 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.3/15, workload match +5.0 | 86.4 tok/s | Runs well |
| Apertus 8B Instruct | 69 ptsGood (B) · 69 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.8/15, workload match +5.0 | 41.0 tok/s | Runs well | 69 ptsGood (B) · 69 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 57.8 tok/s | Runs well |
| Llama 3.1 Nemotron Nano 4B v1.1 | 69 ptsGood (B) · 69 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.9/15, workload match +5.0 | 73.3 tok/s | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15, workload match +5.0 | 47.5 tok/s | Runs well |
| Aya Expanse 8B | 69 ptsGood (B) · 69 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.7/15, workload match +5.0 | 41.1 tok/s | Runs well | 69 ptsGood (B) · 69 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 57.9 tok/s | Runs well |
| Gemma 3 4B IT | 69 ptsGood (B) · 69 of 100 points: Model quality 20.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.4/15, workload match +5.0 | 76.8 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 20.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.7/15, workload match +5.0 | 49.8 tok/s | Runs well |
| Granite 3.0 8B Instruct | 69 ptsGood (B) · 69 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15, workload match +5.0 | 40.4 tok/s | Runs well | 68 ptsGood (B) · 68 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15, workload match +5.0 | 56.9 tok/s | Runs well |
| LFM2.5 2.6B | 69 ptsGood (B) · 69 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.8/15, workload match +5.0 | 56.4 tok/s | Runs well | 68 ptsGood (B) · 68 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.1/15, workload match +5.0 | 79.4 tok/s | Runs well |
| InternLM2.5 7B Chat | 69 ptsGood (B) · 69 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15, workload match +5.0 | 42.7 tok/s | Runs well | 68 ptsGood (B) · 68 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15, workload match +5.0 | 60.1 tok/s | Runs well |
| Nous Hermes 2 SOLAR 10.7B | 69 ptsGood (B) · 69 of 100 points: Model quality 17.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.8/15, workload match +5.0 | 42.0 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.3/15, workload match +5.0 | 43.3 tok/s | Runs well |
| SOLAR 10.7B Instruct | 69 ptsGood (B) · 69 of 100 points: Model quality 17.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.8/15, workload match +5.0 | 42.0 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.3/15, workload match +5.0 | 43.3 tok/s | Runs well |
| Phi-4 Mini (3.8B) | 69 ptsGood (B) · 69 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.6/15, workload match +5.0 | 86.0 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.7/15, workload match +5.0 | 55.8 tok/s | Runs well |
| Ministral 3 3B | 69 ptsGood (B) · 69 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.5/15, workload match +5.0 | 85.8 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15, workload match +5.0 | 55.7 tok/s | Runs well |
| SmolLM3 3B | 69 ptsGood (B) · 69 of 100 points: Model quality 18.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.4/15, workload match +5.0 | 49.4 tok/s | Runs well | 68 ptsGood (B) · 68 of 100 points: Model quality 18.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.7/15, workload match +5.0 | 69.6 tok/s | Runs well |
| DeepSeek R1 Distill Llama 8B | 69 ptsGood (B) · 69 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.7/15 | 41.1 tok/s | Runs well | 68 ptsGood (B) · 68 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15 | 57.9 tok/s | Runs well |
| MedGemma 4B IT | 69 ptsGood (B) · 69 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.4/15, workload match +5.0 | 76.8 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.7/15, workload match +5.0 | 49.8 tok/s | Runs well |
| GLM-4.5 Air (106B-A12B) | 69 ptsGood (B) · 69 of 100 points: Model quality 24.6/45, Memory fit 10.0/25, Speed 14.6/15, Memory use 15.0/15, workload match +5.0 | 37.8 tok/s | Runs with offload | 69 ptsGood (B) · 69 of 100 points: Model quality 24.6/45, Memory fit 10.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 53.5 tok/s | Runs with offload |
| Fara 7B | 69 ptsGood (B) · 69 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15 | 39.9 tok/s | Runs well | 68 ptsGood (B) · 68 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15 | 56.1 tok/s | Runs well |
| Qwen1.5 14B Chat | 69 ptsGood (B) · 69 of 100 points: Model quality 14.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 9.5/15, workload match +5.0 | 45.8 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 15.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 10.3/15, workload match +5.0 | 44.7 tok/s | Runs well |
| Qwen2.5-Math 72B Instruct | 68 ptsGood (B) · 68 of 100 points: Model quality 21.9/45, Memory fit 24.0/25, Speed 8.1/15, Memory use 15.0/15 | 11.5 tok/s | Tight fit | 71 ptsGood (B) · 71 of 100 points: Model quality 21.9/45, Memory fit 25.0/25, Speed 9.6/15, Memory use 15.0/15 | 16.2 tok/s | Runs well |
| Jamba2 3B | 68 ptsGood (B) · 68 of 100 points: Model quality 18.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.1/15, workload match +5.0 | 50.3 tok/s | Runs well | 68 ptsGood (B) · 68 of 100 points: Model quality 18.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.4/15, workload match +5.0 | 70.7 tok/s | Runs well |
| Dolphin 2.9 Llama 3 8B | 68 ptsGood (B) · 68 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.7/15, workload match +5.0 | 41.1 tok/s | Runs well | 68 ptsGood (B) · 68 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 57.9 tok/s | Runs well |
| Hunyuan 4B Instruct | 68 ptsGood (B) · 68 of 100 points: Model quality 18.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.9/15, workload match +5.0 | 78.3 tok/s | Runs well | 69 ptsGood (B) · 69 of 100 points: Model quality 18.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 50.8 tok/s | Runs well |
| Granite 4.0 Micro (3B) | 68 ptsGood (B) · 68 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.7/15, workload match +5.0 | 44.8 tok/s | Runs well | 68 ptsGood (B) · 68 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.0/15, workload match +5.0 | 63.0 tok/s | Runs well |
| Phi-3.5 Mini Instruct | 68 ptsGood (B) · 68 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15, workload match +5.0 | 86.5 tok/s | Runs well | 69 ptsGood (B) · 69 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.7/15, workload match +5.0 | 56.1 tok/s | Runs well |
| DeepSeek LLM 67B Chat | 68 ptsGood (B) · 68 of 100 points: Model quality 15.1/45, Memory fit 25.0/25, Speed 8.4/15, Memory use 15.0/15, workload match +5.0 | 12.4 tok/s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 15.1/45, Memory fit 25.0/25, Speed 9.9/15, Memory use 15.0/15, workload match +5.0 | 17.5 tok/s | Runs well |
| EXAONE 3.0 7.8B Instruct | 68 ptsGood (B) · 68 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.4/15, workload match +5.0 | 42.3 tok/s | Runs well | 67 ptsGood (B) · 67 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.7/15, workload match +5.0 | 59.5 tok/s | Runs well |
| Mistral 7B Instruct v0.3 | 68 ptsGood (B) · 68 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15, workload match +5.0 | 45.6 tok/s | Runs well | 67 ptsGood (B) · 67 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.7/15, workload match +5.0 | 64.1 tok/s | Runs well |
| Zamba2 7B Instruct | 68 ptsGood (B) · 68 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15, workload match +5.0 | 44.9 tok/s | Runs well | 67 ptsGood (B) · 67 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15, workload match +5.0 | 63.2 tok/s | Runs well |
| OLMo 2 7B Instruct | 68 ptsGood (B) · 68 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.9/15, workload match +5.0 | 45.3 tok/s | Runs well | 67 ptsGood (B) · 67 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15, workload match +5.0 | 63.7 tok/s | Runs well |
| Qwen3.5 122B-A10B | 67 ptsGood (B) · 67 of 100 points: Model quality 29.0/45, Memory fit 4.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 45.3 tok/s | Heavy offload | 74 ptsGreat (A) · 74 of 100 points: Model quality 34.8/45, Memory fit 10.0/25, Speed 9.4/15, Memory use 15.0/15, workload match +5.0 | 15.8 tok/s | Runs with offload |
| Falcon3 Mamba 7B Instruct | 67 ptsGood (B) · 67 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15, workload match +5.0 | 45.4 tok/s | Runs well | 67 ptsGood (B) · 67 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.2/15, workload match +5.0 | 64.0 tok/s | Runs well |
| Phi-3.5 Vision Instruct | 67 ptsGood (B) · 67 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.8/15, workload match +5.0 | 79.6 tok/s | Runs well | 69 ptsGood (B) · 69 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.0/15, workload match +5.0 | 51.6 tok/s | Runs well |
| AFM 4.5B | 67 ptsGood (B) · 67 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.8/15, workload match +5.0 | 71.5 tok/s | Runs well | 69 ptsGood (B) · 69 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15, workload match +5.0 | 46.4 tok/s | Runs well |
| DeepSeek MoE 16B Chat | 67 ptsGood (B) · 67 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 10.9/15, workload match +5.0 | 59.0 tok/s | Runs well | 66 ptsGood (B) · 66 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 10.2/15, workload match +5.0 | 83.0 tok/s | Runs well |
| Hermes 2 Pro Mistral 7B | 67 ptsGood (B) · 67 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15, workload match +5.0 | 45.6 tok/s | Runs well | 66 ptsGood (B) · 66 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15, workload match +5.0 | 64.2 tok/s | Runs well |
| Starling LM 7B Beta | 67 ptsGood (B) · 67 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15, workload match +5.0 | 45.6 tok/s | Runs well | 66 ptsGood (B) · 66 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15, workload match +5.0 | 64.2 tok/s | Runs well |
| Qwen3.5 2B | 67 ptsGood (B) · 67 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.4/15, workload match +5.0 | 67.1 tok/s | Runs well | 66 ptsGood (B) · 66 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.7/15, workload match +5.0 | 94.4 tok/s | Runs well |
| Granite 4.0 H Micro | 67 ptsGood (B) · 67 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.2/15, workload match +5.0 | 47.7 tok/s | Runs well | 66 ptsGood (B) · 66 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.5/15, workload match +5.0 | 67.2 tok/s | Runs well |
| DeepSeek Coder 33B Instruct | 67 ptsGood (B) · 67 of 100 points: Model quality 18.8/45, Memory fit 25.0/25, Speed 10.5/15, Memory use 12.9/15 | 19.5 tok/s | Runs well | 68 ptsGood (B) · 68 of 100 points: Model quality 18.8/45, Memory fit 25.0/25, Speed 12.4/15, Memory use 12.2/15 | 27.4 tok/s | Runs well |
| OpenChat 3.5 (0106) | 67 ptsGood (B) · 67 of 100 points: Model quality 15.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15, workload match +5.0 | 45.6 tok/s | Runs well | 66 ptsGood (B) · 66 of 100 points: Model quality 15.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15, workload match +5.0 | 64.2 tok/s | Runs well |
| Aya 23 8B | 67 ptsGood (B) · 67 of 100 points: Model quality 15.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.7/15, workload match +5.0 | 41.1 tok/s | Runs well | 66 ptsGood (B) · 66 of 100 points: Model quality 15.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 57.9 tok/s | Runs well |
| TMax 2B | 66 ptsGood (B) · 66 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.0/15, workload match +5.0 | 81.0 tok/s | Runs well | 66 ptsGood (B) · 66 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.3/15, workload match +5.0 | 114 tok/s | Runs well |
| DeepSeek R1 Distill Qwen 7B | 66 ptsGood (B) · 66 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15 | 43.4 tok/s | Runs well | 66 ptsGood (B) · 66 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15 | 61.0 tok/s | Runs well |
| Llama 2 70B Chat | 66 ptsGood (B) · 66 of 100 points: Model quality 13.6/45, Memory fit 25.0/25, Speed 8.3/15, Memory use 15.0/15, workload match +5.0 | 12.2 tok/s | Runs well | 68 ptsGood (B) · 68 of 100 points: Model quality 13.6/45, Memory fit 25.0/25, Speed 9.8/15, Memory use 15.0/15, workload match +5.0 | 17.1 tok/s | Runs well |
| Yi 1.5 9B Chat | 66 ptsGood (B) · 66 of 100 points: Model quality 16.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.8/15, workload match +5.0 | 51.0 tok/s | Runs well | 67 ptsGood (B) · 67 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 52.7 tok/s | Runs well |
| Phi-3 Mini 128K Instruct | 66 ptsGood (B) · 66 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15, workload match +5.0 | 86.5 tok/s | Runs well | 67 ptsGood (B) · 67 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.7/15, workload match +5.0 | 56.1 tok/s | Runs well |
| Mistral 7B Instruct v0.2 | 66 ptsGood (B) · 66 of 100 points: Model quality 15.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15, workload match +5.0 | 45.6 tok/s | Runs well | 65 ptsGood (B) · 65 of 100 points: Model quality 15.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15, workload match +5.0 | 64.2 tok/s | Runs well |
| Nous Hermes 2 Mistral 7B DPO | 66 ptsGood (B) · 66 of 100 points: Model quality 15.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15, workload match +5.0 | 45.6 tok/s | Runs well | 65 ptsGood (B) · 65 of 100 points: Model quality 15.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15, workload match +5.0 | 64.2 tok/s | Runs well |
| OpenChat 3.5 (1210) | 66 ptsGood (B) · 66 of 100 points: Model quality 15.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15, workload match +5.0 | 45.6 tok/s | Runs well | 65 ptsGood (B) · 65 of 100 points: Model quality 15.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15, workload match +5.0 | 64.2 tok/s | Runs well |
| CodeGeeX4 All 9B | 66 ptsGood (B) · 66 of 100 points: Model quality 20.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.0/15 | 47.9 tok/s | Runs well | 67 ptsGood (B) · 67 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.4/15 | 49.5 tok/s | Runs well |
| Falcon-H1 3B Instruct | 66 ptsGood (B) · 66 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.3/15, workload match +5.0 | 48.3 tok/s | Runs well | 65 ptsGood (B) · 65 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.6/15, workload match +5.0 | 68.0 tok/s | Runs well |
| MiniCPM3 4B | 66 ptsGood (B) · 66 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.3/15, workload match +5.0 | 81.2 tok/s | Runs well | 67 ptsGood (B) · 67 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.5/15, workload match +5.0 | 52.6 tok/s | Runs well |
| Llama 3.2 3B Instruct | 66 ptsGood (B) · 66 of 100 points: Model quality 15.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.7/15, workload match +5.0 | 47.4 tok/s | Runs well | 65 ptsGood (B) · 65 of 100 points: Model quality 15.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.0/15, workload match +5.0 | 66.8 tok/s | Runs well |
| Nemotron 3 Super 120B-A12B | 66 ptsGood (B) · 66 of 100 points: Model quality 27.7/45, Memory fit 4.0/25, Speed 14.6/15, Memory use 15.0/15, workload match +5.0 | 37.8 tok/s | Heavy offload | 72 ptsGreat (A) · 72 of 100 points: Model quality 33.3/45, Memory fit 10.0/25, Speed 8.7/15, Memory use 15.0/15, workload match +5.0 | 13.6 tok/s | Runs with offload |
| Voxtral Mini 3B | 66 ptsGood (B) · 66 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.0/15, workload match +5.0 | 70.6 tok/s | Runs well | 67 ptsGood (B) · 67 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.4/15, workload match +5.0 | 45.8 tok/s | Runs well |
| Qwen2.5-VL 3B Instruct | 66 ptsGood (B) · 66 of 100 points: Model quality 15.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15, workload match +5.0 | 40.6 tok/s | Runs well | 65 ptsGood (B) · 65 of 100 points: Model quality 15.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.2/15, workload match +5.0 | 57.1 tok/s | Runs well |
| Gemma 3n E2B | 66 ptsGood (B) · 66 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.0/15, workload match +5.0 | 60.7 tok/s | Runs well | 65 ptsGood (B) · 65 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.3/15, workload match +5.0 | 85.5 tok/s | Runs well |
| RecurrentGemma 9B IT | 66 ptsGood (B) · 66 of 100 points: Model quality 15.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.0/15, workload match +5.0 | 46.8 tok/s | Runs well | 66 ptsGood (B) · 66 of 100 points: Model quality 15.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.4/15, workload match +5.0 | 48.3 tok/s | Runs well |
| Phi-3 Vision 128K Instruct | 66 ptsGood (B) · 66 of 100 points: Model quality 15.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.8/15, workload match +5.0 | 79.6 tok/s | Runs well | 67 ptsGood (B) · 67 of 100 points: Model quality 15.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.0/15, workload match +5.0 | 51.6 tok/s | Runs well |
| Phi-3 Mini 4K Instruct | 66 ptsGood (B) · 66 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.9/15, workload match +5.0 | 86.5 tok/s | Runs well | 67 ptsGood (B) · 67 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15, workload match +5.0 | 56.1 tok/s | Runs well |
| Qwen1.5 MoE A2.7B Chat | 66 ptsGood (B) · 66 of 100 points: Model quality 10.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 10.2/15, workload match +5.0 | 61.2 tok/s | Runs well | 65 ptsGood (B) · 65 of 100 points: Model quality 10.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 9.5/15, workload match +5.0 | 86.1 tok/s | Runs well |
| Qwen3-VL 2B Instruct | 65 ptsGood (B) · 65 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.6/15, workload match +5.0 | 71.5 tok/s | Runs well | 65 ptsGood (B) · 65 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.9/15, workload match +5.0 | 101 tok/s | Runs well |
| OLMoE 1B-7B Instruct | 65 ptsGood (B) · 65 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 9.0/15, workload match +5.0 | 58.6 tok/s | Runs well | 65 ptsGood (B) · 65 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.4/15, workload match +5.0 | 82.4 tok/s | Runs well |
| OpenHermes 2.5 Mistral 7B | 65 ptsGood (B) · 65 of 100 points: Model quality 14.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15, workload match +5.0 | 45.6 tok/s | Runs well | 65 ptsGood (B) · 65 of 100 points: Model quality 14.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15, workload match +5.0 | 64.2 tok/s | Runs well |
| Starling LM 7B Alpha | 65 ptsGood (B) · 65 of 100 points: Model quality 14.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15, workload match +5.0 | 45.6 tok/s | Runs well | 65 ptsGood (B) · 65 of 100 points: Model quality 14.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15, workload match +5.0 | 64.2 tok/s | Runs well |
| DeepSeek Prover V2 7B | 65 ptsGood (B) · 65 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.6/15 | 47.8 tok/s | Runs well | 64 ptsGood (B) · 64 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.9/15 | 67.3 tok/s | Runs well |
| Qwen2.5 3B Instruct | 65 ptsGood (B) · 65 of 100 points: Model quality 15.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.3/15, workload match +5.0 | 49.3 tok/s | Runs well | 64 ptsGood (B) · 64 of 100 points: Model quality 15.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.6/15, workload match +5.0 | 69.3 tok/s | Runs well |
| Orca 2 13B | 65 ptsGood (B) · 65 of 100 points: Model quality 12.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.2/15, workload match +5.0 | 41.2 tok/s | Runs well | 65 ptsGood (B) · 65 of 100 points: Model quality 12.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.2/15, workload match +5.0 | 48.7 tok/s | Runs well |
| Phi-4 Mini Reasoning (3.8B) | 65 ptsGood (B) · 65 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.6/15 | 86.0 tok/s | Runs well | 66 ptsGood (B) · 66 of 100 points: Model quality 20.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.7/15 | 55.8 tok/s | Runs well |
| Code Llama 70B Instruct | 65 ptsGood (B) · 65 of 100 points: Model quality 18.8/45, Memory fit 24.0/25, Speed 7.3/15, Memory use 15.0/15 | 9.4 tok/s | Tight fit | 66 ptsGood (B) · 66 of 100 points: Model quality 16.6/45, Memory fit 25.0/25, Speed 9.8/15, Memory use 15.0/15 | 17.1 tok/s | Runs well |
| Hermes 3 Llama 3.2 3B | 65 ptsGood (B) · 65 of 100 points: Model quality 14.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.7/15, workload match +5.0 | 47.4 tok/s | Runs well | 64 ptsGood (B) · 64 of 100 points: Model quality 14.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.0/15, workload match +5.0 | 66.8 tok/s | Runs well |
| LFM2 2.6B | 64 ptsGood (B) · 64 of 100 points: Model quality 15.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.7/15, workload match +5.0 | 59.2 tok/s | Runs well | 64 ptsGood (B) · 64 of 100 points: Model quality 15.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.0/15, workload match +5.0 | 83.4 tok/s | Runs well |
| Llama 4 Scout 17B-16E | 64 ptsGood (B) · 64 of 100 points: Model quality 22.7/45, Memory fit 10.0/25, Speed 12.2/15, Memory use 15.0/15, workload match +5.0 | 26.6 tok/s | Runs with offload | 67 ptsGood (B) · 67 of 100 points: Model quality 22.7/45, Memory fit 10.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 43.6 tok/s | Runs with offload |
| Mistral 7B Instruct v0.1 | 64 ptsGood (B) · 64 of 100 points: Model quality 13.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15, workload match +5.0 | 45.6 tok/s | Runs well | 64 ptsGood (B) · 64 of 100 points: Model quality 13.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15, workload match +5.0 | 64.2 tok/s | Runs well |
| Qwen1.5 7B Chat | 64 ptsGood (B) · 64 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.1/15, workload match +5.0 | 42.8 tok/s | Runs well | 64 ptsGood (B) · 64 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.4/15, workload match +5.0 | 60.2 tok/s | Runs well |
| Ling 3.0 Flash (127B-A5B) | 64 ptsGood (B) · 64 of 100 points: Model quality 25.8/45, Memory fit 4.0/25, Speed 15.0/15, Memory use 15.0/15, workload match +5.0 | 88.8 tok/s | Heavy offload | 73 ptsGreat (A) · 73 of 100 points: Model quality 31.0/45, Memory fit 10.0/25, Speed 12.9/15, Memory use 15.0/15, workload match +5.0 | 29.6 tok/s | Runs with offload |
| Gemma 1.1 7B IT | 64 ptsGood (B) · 64 of 100 points: Model quality 12.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.3/15, workload match +5.0 | 52.7 tok/s | Runs well | 65 ptsGood (B) · 65 of 100 points: Model quality 12.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.6/15, workload match +5.0 | 54.4 tok/s | Runs well |
| Granite Code 34B Instruct | 64 ptsGood (B) · 64 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 10.4/15, Memory use 12.2/15 | 19.2 tok/s | Runs well | 65 ptsGood (B) · 65 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 12.3/15, Memory use 11.5/15 | 27.1 tok/s | Runs well |
| Llama 2 13B Chat | 64 ptsGood (B) · 64 of 100 points: Model quality 11.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.2/15, workload match +5.0 | 41.2 tok/s | Runs well | 64 ptsGood (B) · 64 of 100 points: Model quality 11.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.2/15, workload match +5.0 | 48.7 tok/s | Runs well |
| Code Llama 34B Instruct | 64 ptsGood (B) · 64 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 10.4/15, Memory use 12.8/15 | 19.2 tok/s | Runs well | 65 ptsGood (B) · 65 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 12.3/15, Memory use 12.1/15 | 27.1 tok/s | Runs well |
| InternLM2 Math Plus 20B | 64 ptsGood (B) · 64 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 13.6/15, Memory use 8.7/15 | 32.7 tok/s | Runs well | 65 ptsGood (B) · 65 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.0/15 | 46.0 tok/s | Runs well |
| Falcon Mamba 7B Instruct | 64 ptsGood (B) · 64 of 100 points: Model quality 13.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15, workload match +5.0 | 45.4 tok/s | Runs well | 63 ptsGood (B) · 63 of 100 points: Model quality 13.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.2/15, workload match +5.0 | 64.0 tok/s | Runs well |
| StarCoder2 15B Instruct | 64 ptsGood (B) · 64 of 100 points: Model quality 17.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.1/15 | 40.7 tok/s | Runs well | 65 ptsGood (B) · 65 of 100 points: Model quality 17.8/45, Memory fit 25.0/25, Speed 14.9/15, Memory use 8.1/15 | 39.7 tok/s | Runs well |
| LLaVA 1.6 Mistral 7B | 64 ptsGood (B) · 64 of 100 points: Model quality 12.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.5/15, workload match +5.0 | 43.6 tok/s | Runs well | 63 ptsGood (B) · 63 of 100 points: Model quality 12.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.8/15, workload match +5.0 | 61.4 tok/s | Runs well |
| Falcon 3 3B Instruct | 64 ptsGood (B) · 64 of 100 points: Model quality 13.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15, workload match +5.0 | 47.1 tok/s | Runs well | 63 ptsGood (B) · 63 of 100 points: Model quality 13.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.9/15, workload match +5.0 | 66.3 tok/s | Runs well |
| Neural Chat 7B v3.3 | 63 ptsGood (B) · 63 of 100 points: Model quality 12.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15, workload match +5.0 | 45.6 tok/s | Runs well | 63 ptsGood (B) · 63 of 100 points: Model quality 12.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15, workload match +5.0 | 64.2 tok/s | Runs well |
| Baichuan 2 13B Chat | 63 ptsGood (B) · 63 of 100 points: Model quality 11.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.8/15, workload match +5.0 | 46.7 tok/s | Runs well | 65 ptsGood (B) · 65 of 100 points: Model quality 11.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.5/15, workload match +5.0 | 45.6 tok/s | Runs well |
| StableLM 2 12B Chat | 63 ptsGood (B) · 63 of 100 points: Model quality 12.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.7/15, workload match +5.0 | 44.2 tok/s | Runs well | 64 ptsGood (B) · 64 of 100 points: Model quality 12.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15, workload match +5.0 | 52.2 tok/s | Runs well |
| Gemma 7B IT | 63 ptsGood (B) · 63 of 100 points: Model quality 11.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.3/15, workload match +5.0 | 52.7 tok/s | Runs well | 64 ptsGood (B) · 64 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.6/15, workload match +5.0 | 54.4 tok/s | Runs well |
| Yi Coder 9B Chat | 63 ptsGood (B) · 63 of 100 points: Model quality 17.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.0/15 | 51.0 tok/s | Runs well | 64 ptsGood (B) · 64 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15 | 52.7 tok/s | Runs well |
| Qwen2.5-Omni 3B | 63 ptsGood (B) · 63 of 100 points: Model quality 13.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.1/15, workload match +5.0 | 59.6 tok/s | Runs well | 62 ptsGood (B) · 62 of 100 points: Model quality 13.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.4/15, workload match +5.0 | 83.9 tok/s | Runs well |
| Unlimited OCR (3B-A0.6B) | 63 ptsGood (B) · 63 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15 | 127 tok/s | Runs well | 62 ptsGood (B) · 62 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.9/15 | 179 tok/s | Runs well |
| Granite 3.2 2B Instruct | 63 ptsGood (B) · 63 of 100 points: Model quality 13.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.9/15, workload match +5.0 | 60.2 tok/s | Runs well | 62 ptsGood (B) · 62 of 100 points: Model quality 13.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.2/15, workload match +5.0 | 84.7 tok/s | Runs well |
| Granite 3.3 2B | 63 ptsGood (B) · 63 of 100 points: Model quality 13.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.9/15, workload match +5.0 | 60.2 tok/s | Runs well | 62 ptsGood (B) · 62 of 100 points: Model quality 13.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.2/15, workload match +5.0 | 84.7 tok/s | Runs well |
| Leanstral 119B-A6B | 63 ptsGood (B) · 63 of 100 points: Model quality 23.3/45, Memory fit 10.0/25, Speed 15.0/15, Memory use 15.0/15 | 69.7 tok/s | Runs with offload | 65 ptsGood (B) · 65 of 100 points: Model quality 28.0/45, Memory fit 10.0/25, Speed 12.3/15, Memory use 15.0/15 | 26.9 tok/s | Runs with offload |
| EXAONE 3.5 2.4B | 63 ptsGood (B) · 63 of 100 points: Model quality 13.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.8/15, workload match +5.0 | 63.2 tok/s | Runs well | 62 ptsGood (B) · 62 of 100 points: Model quality 13.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.1/15, workload match +5.0 | 88.9 tok/s | Runs well |
| Zephyr 7B Beta | 63 ptsGood (B) · 63 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15, workload match +5.0 | 45.6 tok/s | Runs well | 62 ptsGood (B) · 62 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15, workload match +5.0 | 64.2 tok/s | Runs well |
| Qwen3 1.7B | 63 ptsGood (B) · 63 of 100 points: Model quality 13.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.5/15, workload match +5.0 | 75.0 tok/s | Runs well | 62 ptsGood (B) · 62 of 100 points: Model quality 13.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.8/15, workload match +5.0 | 106 tok/s | Runs well |
| Yi 1.5 6B Chat | 62 ptsGood (B) · 62 of 100 points: Model quality 12.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.4/15, workload match +5.0 | 54.5 tok/s | Runs well | 62 ptsGood (B) · 62 of 100 points: Model quality 12.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.7/15, workload match +5.0 | 76.7 tok/s | Runs well |
| DeepSeek Coder 6.7B Instruct | 62 ptsGood (B) · 62 of 100 points: Model quality 15.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.6/15 | 49.0 tok/s | Runs well | 62 ptsGood (B) · 62 of 100 points: Model quality 15.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.9/15 | 69.0 tok/s | Runs well |
| LLaVA 1.5 13B | 62 ptsGood (B) · 62 of 100 points: Model quality 9.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.3/15, workload match +5.0 | 40.2 tok/s | Runs well | 63 ptsGood (B) · 63 of 100 points: Model quality 9.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.3/15, workload match +5.0 | 47.5 tok/s | Runs well |
| Code Llama 13B Instruct | 62 ptsGood (B) · 62 of 100 points: Model quality 13.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 9.8/15 | 41.2 tok/s | Runs well | 63 ptsGood (B) · 63 of 100 points: Model quality 13.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 9.8/15 | 48.7 tok/s | Runs well |
| WizardLM 13B v1.2 | 62 ptsGood (B) · 62 of 100 points: Model quality 9.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.2/15, workload match +5.0 | 41.2 tok/s | Runs well | 62 ptsGood (B) · 62 of 100 points: Model quality 9.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.2/15, workload match +5.0 | 48.7 tok/s | Runs well |
| LFM2.5 1.2B Instruct | 62 ptsGood (B) · 62 of 100 points: Model quality 14.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.3/15, workload match +5.0 | 130 tok/s | Runs well | 62 ptsGood (B) · 62 of 100 points: Model quality 14.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.6/15, workload match +5.0 | 183 tok/s | Runs well |
| Granite 3.1 2B Instruct | 62 ptsGood (B) · 62 of 100 points: Model quality 12.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.9/15, workload match +5.0 | 60.2 tok/s | Runs well | 61 ptsGood (B) · 61 of 100 points: Model quality 12.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.2/15, workload match +5.0 | 84.7 tok/s | Runs well |
| Gemma 2 2B | 62 ptsGood (B) · 62 of 100 points: Model quality 12.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.9/15, workload match +5.0 | 58.3 tok/s | Runs well | 61 ptsGood (B) · 61 of 100 points: Model quality 12.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.2/15, workload match +5.0 | 82.1 tok/s | Runs well |
| CodeGemma 7B | 62 ptsGood (B) · 62 of 100 points: Model quality 15.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.3/15 | 52.7 tok/s | Runs well | 62 ptsGood (B) · 62 of 100 points: Model quality 15.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.6/15 | 54.4 tok/s | Runs well |
| Qwen2.5-Math 7B Instruct | 62 ptsGood (B) · 62 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15 | 43.4 tok/s | Runs well | 61 ptsGood (B) · 61 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.4/15 | 61.0 tok/s | Runs well |
| EXAONE 4.0 1.2B | 62 ptsGood (B) · 62 of 100 points: Model quality 13.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.6/15, workload match +5.0 | 119 tok/s | Runs well | 61 ptsGood (B) · 61 of 100 points: Model quality 13.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.9/15, workload match +5.0 | 167 tok/s | Runs well |
| Codestral Mamba 7B | 62 ptsGood (B) · 62 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15 | 45.3 tok/s | Runs well | 61 ptsGood (B) · 61 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.2/15 | 63.8 tok/s | Runs well |
| Granite Vision 3.2 2B | 62 ptsGood (B) · 62 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.3/15, workload match +5.0 | 51.1 tok/s | Runs well | 61 ptsGood (B) · 61 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.6/15, workload match +5.0 | 71.9 tok/s | Runs well |
| Granite Code 20B Instruct | 62 ptsGood (B) · 62 of 100 points: Model quality 15.4/45, Memory fit 25.0/25, Speed 13.5/15, Memory use 8.1/15 | 32.3 tok/s | Runs well | 62 ptsGood (B) · 62 of 100 points: Model quality 15.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.4/15 | 45.5 tok/s | Runs well |
| Command A (111B) | 61 ptsGood (B) · 61 of 100 points: Model quality 25.2/45, Memory fit 10.0/25, Speed 6.8/15, Memory use 15.0/15, workload match +5.0 | 8.2 tok/s | Runs with offload | 64 ptsGood (B) · 64 of 100 points: Model quality 30.2/45, Memory fit 10.0/25, Speed 4.5/15, Memory use 15.0/15, workload match +5.0 | 3.7 tok/s | Runs with offload |
| Vicuna 13B v1.5 | 61 ptsGood (B) · 61 of 100 points: Model quality 8.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.2/15, workload match +5.0 | 41.2 tok/s | Runs well | 62 ptsGood (B) · 62 of 100 points: Model quality 8.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.2/15, workload match +5.0 | 48.7 tok/s | Runs well |
| Mathstral 7B | 61 ptsGood (B) · 61 of 100 points: Model quality 15.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15 | 45.6 tok/s | Runs well | 60 ptsGood (B) · 60 of 100 points: Model quality 15.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.7/15 | 64.1 tok/s | Runs well |
| Granite 3.0 2B Instruct | 61 ptsGood (B) · 61 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.8/15, workload match +5.0 | 57.9 tok/s | Runs well | 60 ptsGood (B) · 60 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.1/15, workload match +5.0 | 81.5 tok/s | Runs well |
| Orca 2 7B | 61 ptsGood (B) · 61 of 100 points: Model quality 9.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15, workload match +5.0 | 49.0 tok/s | Runs well | 60 ptsGood (B) · 60 of 100 points: Model quality 9.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15, workload match +5.0 | 69.0 tok/s | Runs well |
| Qwen2.5 Coder 3B Instruct | 61 ptsGood (B) · 61 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.3/15 | 49.3 tok/s | Runs well | 60 ptsGood (B) · 60 of 100 points: Model quality 16.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.6/15 | 69.3 tok/s | Runs well |
| Granite Code 8B Instruct (128K) | 61 ptsGood (B) · 61 of 100 points: Model quality 14.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.8/15 | 41.0 tok/s | Runs well | 60 ptsGood (B) · 60 of 100 points: Model quality 14.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15 | 57.8 tok/s | Runs well |
| Granite 4.0 H 1B | 61 ptsGood (B) · 61 of 100 points: Model quality 12.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.6/15, workload match +5.0 | 104 tok/s | Runs well | 60 ptsGood (B) · 60 of 100 points: Model quality 12.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.9/15, workload match +5.0 | 147 tok/s | Runs well |
| SQLCoder 7B v2 | 61 ptsGood (B) · 61 of 100 points: Model quality 13.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.6/15 | 49.0 tok/s | Runs well | 60 ptsGood (B) · 60 of 100 points: Model quality 13.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.9/15 | 69.0 tok/s | Runs well |
| Laguna S 2.1 (118B-A8B) | 61 ptsGood (B) · 61 of 100 points: Model quality 27.1/45, Memory fit 4.0/25, Speed 15.0/15, Memory use 15.0/15 | 56.6 tok/s | Heavy offload | 68 ptsGood (B) · 68 of 100 points: Model quality 32.5/45, Memory fit 10.0/25, Speed 10.9/15, Memory use 15.0/15 | 21.0 tok/s | Runs with offload |
| InternLM2 Math Plus 7B | 60 ptsGood (B) · 60 of 100 points: Model quality 14.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15 | 42.7 tok/s | Runs well | 60 ptsGood (B) · 60 of 100 points: Model quality 14.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15 | 60.1 tok/s | Runs well |
| Baichuan 2 7B Chat | 60 ptsGood (B) · 60 of 100 points: Model quality 9.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.0/15, workload match +5.0 | 44.0 tok/s | Runs well | 60 ptsGood (B) · 60 of 100 points: Model quality 9.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15, workload match +5.0 | 61.9 tok/s | Runs well |
| SmolVLM2 2.2B Instruct | 60 ptsGood (B) · 60 of 100 points: Model quality 10.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.1/15, workload match +5.0 | 67.7 tok/s | Runs well | 60 ptsGood (B) · 60 of 100 points: Model quality 10.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.4/15, workload match +5.0 | 95.2 tok/s | Runs well |
| StarCoder2 15B (base) | 60 ptsGood (B) · 60 of 100 points: Model quality 13.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.1/15 | 40.7 tok/s | Runs well | 62 ptsGood (B) · 62 of 100 points: Model quality 14.3/45, Memory fit 25.0/25, Speed 14.9/15, Memory use 8.1/15 | 39.7 tok/s | Runs well |
| Llama 2 7B Chat | 60 ptsGood (B) · 60 of 100 points: Model quality 9.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15, workload match +5.0 | 49.0 tok/s | Runs well | 59 ptsGood (B) · 59 of 100 points: Model quality 9.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15, workload match +5.0 | 69.0 tok/s | Runs well |
| Falcon-H1 1.5B Instruct | 60 ptsGood (B) · 60 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.7/15, workload match +5.0 | 98.2 tok/s | Runs well | 59 ptsGood (B) · 59 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.0/15, workload match +5.0 | 138 tok/s | Runs well |
| Qwen2-VL 2B Instruct | 60 ptsGood (B) · 60 of 100 points: Model quality 10.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.4/15, workload match +5.0 | 68.9 tok/s | Runs well | 59 ptsGood (B) · 59 of 100 points: Model quality 10.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.7/15, workload match +5.0 | 97.0 tok/s | Runs well |
| LFM2 1.2B | 60 ptsGood (B) · 60 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.3/15, workload match +5.0 | 130 tok/s | Runs well | 59 ptsGood (B) · 59 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.6/15, workload match +5.0 | 183 tok/s | Runs well |
| SmolVLM 2.2B Instruct | 59 ptsGood (B) · 59 of 100 points: Model quality 9.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.1/15, workload match +5.0 | 67.7 tok/s | Runs well | 59 ptsGood (B) · 59 of 100 points: Model quality 9.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.4/15, workload match +5.0 | 95.2 tok/s | Runs well |
| Zamba2 2.7B Instruct | 59 ptsGood (B) · 59 of 100 points: Model quality 9.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.9/15, workload match +5.0 | 57.2 tok/s | Runs well | 59 ptsGood (B) · 59 of 100 points: Model quality 9.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.2/15, workload match +5.0 | 80.6 tok/s | Runs well |
| DeepSeek LLM 7B Chat | 59 ptsGood (B) · 59 of 100 points: Model quality 8.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15, workload match +5.0 | 47.8 tok/s | Runs well | 59 ptsGood (B) · 59 of 100 points: Model quality 8.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15, workload match +5.0 | 67.3 tok/s | Runs well |
| Qwen3.5 0.8B | 59 ptsGood (B) · 59 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.0/15, workload match +5.0 | 175 tok/s | Runs well | 59 ptsGood (B) · 59 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.3/15, workload match +5.0 | 246 tok/s | Runs well |
| Granite 3.1 3B-A800M Instruct | 59 ptsGood (B) · 59 of 100 points: Model quality 9.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15, workload match +5.0 | 95.2 tok/s | Runs well | 58 ptsGood (B) · 58 of 100 points: Model quality 9.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.9/15, workload match +5.0 | 134 tok/s | Runs well |
| Qwen2.5 1.5B Instruct | 59 ptsGood (B) · 59 of 100 points: Model quality 10.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.7/15, workload match +5.0 | 98.9 tok/s | Runs well | 58 ptsGood (B) · 58 of 100 points: Model quality 10.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.0/15, workload match +5.0 | 139 tok/s | Runs well |
| LFM2-VL 1.6B | 59 ptsGood (B) · 59 of 100 points: Model quality 10.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.7/15, workload match +5.0 | 96.4 tok/s | Runs well | 58 ptsGood (B) · 58 of 100 points: Model quality 10.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.0/15, workload match +5.0 | 136 tok/s | Runs well |
| Nemotron Mini 4B Instruct | 59 ptsGood (B) · 59 of 100 points: Model quality 9.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.5/15, workload match +5.0 | 78.9 tok/s | Runs well | 60 ptsGood (B) · 60 of 100 points: Model quality 9.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.7/15, workload match +5.0 | 51.1 tok/s | Runs well |
| Yi 6B Chat | 59 ptsGood (B) · 59 of 100 points: Model quality 9.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.4/15, workload match +5.0 | 54.5 tok/s | Runs well | 58 ptsGood (B) · 58 of 100 points: Model quality 9.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.7/15, workload match +5.0 | 76.7 tok/s | Runs well |
| Command R+ (08-2024) 104B | 59 ptsGood (B) · 59 of 100 points: Model quality 22.1/45, Memory fit 10.0/25, Speed 7.0/15, Memory use 15.0/15, workload match +5.0 | 8.7 tok/s | Runs with offload | 75 ptsGreat (A) · 75 of 100 points: Model quality 22.1/45, Memory fit 24.0/25, Speed 9.0/15, Memory use 15.0/15, workload match +5.0 | 14.4 tok/s | Tight fit |
| Hunyuan 1.8B Instruct | 59 ptsGood (B) · 59 of 100 points: Model quality 9.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.1/15, workload match +5.0 | 85.1 tok/s | Runs well | 58 ptsGood (B) · 58 of 100 points: Model quality 9.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.4/15, workload match +5.0 | 120 tok/s | Runs well |
| LLaVA 1.5 7B | 58 ptsGood (B) · 58 of 100 points: Model quality 7.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.8/15, workload match +5.0 | 46.8 tok/s | Runs well | 58 ptsGood (B) · 58 of 100 points: Model quality 7.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15, workload match +5.0 | 65.9 tok/s | Runs well |
| Granite 3.0 3B-A800M Instruct | 58 ptsGood (B) · 58 of 100 points: Model quality 8.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.5/15, workload match +5.0 | 95.2 tok/s | Runs well | 57 ptsUsable (C) · 57 of 100 points: Model quality 8.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.8/15, workload match +5.0 | 134 tok/s | Runs well |
| SmolLM2 1.7B Instruct | 58 ptsGood (B) · 58 of 100 points: Model quality 9.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.6/15, workload match +5.0 | 89.0 tok/s | Runs well | 57 ptsUsable (C) · 57 of 100 points: Model quality 9.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.9/15, workload match +5.0 | 125 tok/s | Runs well |
| Code Llama 7B Instruct | 58 ptsGood (B) · 58 of 100 points: Model quality 10.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.6/15 | 49.0 tok/s | Runs well | 57 ptsUsable (C) · 57 of 100 points: Model quality 10.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.9/15 | 69.0 tok/s | Runs well |
| Llama 3.2 1B Instruct | 58 ptsGood (B) · 58 of 100 points: Model quality 9.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.5/15, workload match +5.0 | 123 tok/s | Runs well | 57 ptsUsable (C) · 57 of 100 points: Model quality 9.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.8/15, workload match +5.0 | 173 tok/s | Runs well |
| WizardMath 7B v1.1 | 58 ptsGood (B) · 58 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.3/15 | 45.6 tok/s | Runs well | 57 ptsUsable (C) · 57 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.6/15 | 64.2 tok/s | Runs well |
| StarCoder 15B | 57 ptsUsable (C) · 57 of 100 points: Model quality 11.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.9/15 | 41.0 tok/s | Runs well | 59 ptsGood (B) · 59 of 100 points: Model quality 11.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 7.8/15 | 40.1 tok/s | Runs well |
| Vicuna 7B v1.5 | 57 ptsUsable (C) · 57 of 100 points: Model quality 6.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.6/15, workload match +5.0 | 49.0 tok/s | Runs well | 57 ptsUsable (C) · 57 of 100 points: Model quality 6.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.9/15, workload match +5.0 | 69.0 tok/s | Runs well |
| StableLM Zephyr 3B | 57 ptsUsable (C) · 57 of 100 points: Model quality 7.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.5/15, workload match +5.0 | 54.4 tok/s | Runs well | 56 ptsUsable (C) · 56 of 100 points: Model quality 7.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.8/15, workload match +5.0 | 76.5 tok/s | Runs well |
| Command A Reasoning (111B) | 57 ptsUsable (C) · 57 of 100 points: Model quality 25.8/45, Memory fit 10.0/25, Speed 6.8/15, Memory use 15.0/15 | 8.2 tok/s | Runs with offload | 60 ptsGood (B) · 60 of 100 points: Model quality 31.0/45, Memory fit 10.0/25, Speed 4.5/15, Memory use 15.0/15 | 3.7 tok/s | Runs with offload |
| OLMo 7B Instruct | 57 ptsUsable (C) · 57 of 100 points: Model quality 6.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15, workload match +5.0 | 48.0 tok/s | Runs well | 56 ptsUsable (C) · 56 of 100 points: Model quality 6.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.5/15, workload match +5.0 | 67.5 tok/s | Runs well |
| InternLM2.5 1.8B Chat | 57 ptsUsable (C) · 57 of 100 points: Model quality 8.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.3/15, workload match +5.0 | 80.6 tok/s | Runs well | 56 ptsUsable (C) · 56 of 100 points: Model quality 8.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.6/15, workload match +5.0 | 113 tok/s | Runs well |
| Phi-2 2.7B | 57 ptsUsable (C) · 57 of 100 points: Model quality 7.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.1/15, workload match +5.0 | 54.8 tok/s | Runs well | 56 ptsUsable (C) · 56 of 100 points: Model quality 7.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.4/15, workload match +5.0 | 77.1 tok/s | Runs well |
| Falcon 40B Instruct | 57 ptsUsable (C) · 57 of 100 points: Model quality 5.0/45, Memory fit 25.0/25, Speed 12.0/15, Memory use 10.3/15, workload match +5.0 | 25.5 tok/s | Runs well | 58 ptsGood (B) · 58 of 100 points: Model quality 5.0/45, Memory fit 25.0/25, Speed 14.2/15, Memory use 9.6/15, workload match +5.0 | 35.9 tok/s | Runs well |
| Command R+ 104B (04-2024) | 57 ptsUsable (C) · 57 of 100 points: Model quality 20.2/45, Memory fit 10.0/25, Speed 7.0/15, Memory use 15.0/15, workload match +5.0 | 8.7 tok/s | Runs with offload | 73 ptsGreat (A) · 73 of 100 points: Model quality 20.2/45, Memory fit 24.0/25, Speed 9.0/15, Memory use 15.0/15, workload match +5.0 | 14.4 tok/s | Tight fit |
| Gemma 3 1B | 57 ptsUsable (C) · 57 of 100 points: Model quality 9.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.1/15, workload match +5.0 | 152 tok/s | Runs well | 56 ptsUsable (C) · 56 of 100 points: Model quality 9.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.4/15, workload match +5.0 | 214 tok/s | Runs well |
| StarCoder2 7B | 56 ptsUsable (C) · 56 of 100 points: Model quality 10.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.0/15 | 46.1 tok/s | Runs well | 56 ptsUsable (C) · 56 of 100 points: Model quality 10.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.4/15 | 64.8 tok/s | Runs well |
| InternLM 7B | 56 ptsUsable (C) · 56 of 100 points: Model quality 5.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.4/15, workload match +5.0 | 45.1 tok/s | Runs well | 56 ptsUsable (C) · 56 of 100 points: Model quality 5.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.7/15, workload match +5.0 | 63.5 tok/s | Runs well |
| Falcon 3 1B Instruct | 56 ptsUsable (C) · 56 of 100 points: Model quality 7.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.0/15, workload match +5.0 | 91.2 tok/s | Runs well | 55 ptsUsable (C) · 55 of 100 points: Model quality 7.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.3/15, workload match +5.0 | 128 tok/s | Runs well |
| LFM2 700M | 55 ptsUsable (C) · 55 of 100 points: Model quality 8.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.9/15, workload match +5.0 | 206 tok/s | Runs well | 55 ptsUsable (C) · 55 of 100 points: Model quality 8.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.2/15, workload match +5.0 | 290 tok/s | Runs well |
| Qwen2 1.5B Instruct | 55 ptsUsable (C) · 55 of 100 points: Model quality 7.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.7/15, workload match +5.0 | 98.9 tok/s | Runs well | 55 ptsUsable (C) · 55 of 100 points: Model quality 7.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.0/15, workload match +5.0 | 139 tok/s | Runs well |
| Granite Code 3B Instruct (128K) | 55 ptsUsable (C) · 55 of 100 points: Model quality 9.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.8/15 | 43.8 tok/s | Runs well | 55 ptsUsable (C) · 55 of 100 points: Model quality 9.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15 | 61.6 tok/s | Runs well |
| DeepScaleR 1.5B Preview | 55 ptsUsable (C) · 55 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.0/15 | 85.5 tok/s | Runs well | 54 ptsUsable (C) · 54 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.3/15 | 120 tok/s | Runs well |
| Mistral Large 2.1 (2411) | 55 ptsUsable (C) · 55 of 100 points: Model quality 25.2/45, Memory fit 4.0/25, Speed 6.4/15, Memory use 15.0/15, workload match +5.0 | 7.4 tok/s | Heavy offload | 60 ptsGood (B) · 60 of 100 points: Model quality 25.2/45, Memory fit 10.0/25, Speed 5.3/15, Memory use 15.0/15, workload match +5.0 | 5.0 tok/s | Runs with offload |
| Mistral Medium 3.5 (128B) | 55 ptsUsable (C) · 55 of 100 points: Model quality 25.3/45, Memory fit 4.0/25, Speed 6.3/15, Memory use 15.0/15, workload match +5.0 | 7.1 tok/s | Heavy offload | 60 ptsGood (B) · 60 of 100 points: Model quality 25.3/45, Memory fit 10.0/25, Speed 4.9/15, Memory use 15.0/15, workload match +5.0 | 4.2 tok/s | Runs with offload |
| Qwen2.5 Coder 1.5B Instruct | 55 ptsUsable (C) · 55 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.7/15 | 98.9 tok/s | Runs well | 54 ptsUsable (C) · 54 of 100 points: Model quality 11.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.0/15 | 139 tok/s | Runs well |
| SmolLM 1.7B Instruct | 55 ptsUsable (C) · 55 of 100 points: Model quality 6.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.0/15, workload match +5.0 | 89.0 tok/s | Runs well | 54 ptsUsable (C) · 54 of 100 points: Model quality 6.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.3/15, workload match +5.0 | 125 tok/s | Runs well |
| Stable Code Instruct 3B | 55 ptsUsable (C) · 55 of 100 points: Model quality 9.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15 | 54.4 tok/s | Runs well | 54 ptsUsable (C) · 54 of 100 points: Model quality 9.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.4/15 | 76.5 tok/s | Runs well |
| OLMo 2 1B Instruct | 55 ptsUsable (C) · 55 of 100 points: Model quality 6.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.8/15, workload match +5.0 | 103 tok/s | Runs well | 54 ptsUsable (C) · 54 of 100 points: Model quality 6.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.1/15, workload match +5.0 | 145 tok/s | Runs well |
| Gemma 2B IT | 55 ptsUsable (C) · 55 of 100 points: Model quality 5.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.6/15, workload match +5.0 | 60.7 tok/s | Runs well | 54 ptsUsable (C) · 54 of 100 points: Model quality 5.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.9/15, workload match +5.0 | 85.4 tok/s | Runs well |
| Mistral Large 2 (123B) | 55 ptsUsable (C) · 55 of 100 points: Model quality 24.6/45, Memory fit 4.0/25, Speed 6.4/15, Memory use 15.0/15, workload match +5.0 | 7.4 tok/s | Heavy offload | 59 ptsGood (B) · 59 of 100 points: Model quality 24.6/45, Memory fit 10.0/25, Speed 5.3/15, Memory use 15.0/15, workload match +5.0 | 4.9 tok/s | Runs with offload |
| Pixtral Large 124B (2411) | 54 ptsUsable (C) · 54 of 100 points: Model quality 24.6/45, Memory fit 4.0/25, Speed 6.4/15, Memory use 15.0/15, workload match +5.0 | 7.3 tok/s | Heavy offload | 59 ptsGood (B) · 59 of 100 points: Model quality 24.6/45, Memory fit 10.0/25, Speed 5.2/15, Memory use 15.0/15, workload match +5.0 | 4.7 tok/s | Runs with offload |
| Qwen3 0.6B | 54 ptsUsable (C) · 54 of 100 points: Model quality 6.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.3/15, workload match +5.0 | 203 tok/s | Runs well | 53 ptsUsable (C) · 53 of 100 points: Model quality 6.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.6/15, workload match +5.0 | 286 tok/s | Runs well |
| Falcon-H1 0.5B Instruct | 54 ptsUsable (C) · 54 of 100 points: Model quality 6.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.7/15, workload match +5.0 | 293 tok/s | Runs well | 53 ptsUsable (C) · 53 of 100 points: Model quality 6.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.0/15, workload match +5.0 | 412 tok/s | Runs well |
| DeepSeek R1 Distill Qwen 1.5B | 53 ptsUsable (C) · 53 of 100 points: Model quality 9.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.0/15 | 85.5 tok/s | Runs well | 53 ptsUsable (C) · 53 of 100 points: Model quality 9.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.3/15 | 120 tok/s | Runs well |
| LFM2.5 350M | 53 ptsUsable (C) · 53 of 100 points: Model quality 6.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.5/15, workload match +5.0 | 435 tok/s | Runs well | 53 ptsUsable (C) · 53 of 100 points: Model quality 6.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.8/15, workload match +5.0 | 612 tok/s | Runs well |
| StableLM 2 1.6B Chat | 53 ptsUsable (C) · 53 of 100 points: Model quality 4.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.1/15, workload match +5.0 | 92.8 tok/s | Runs well | 52 ptsUsable (C) · 52 of 100 points: Model quality 4.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.4/15, workload match +5.0 | 131 tok/s | Runs well |
| StableLM 2 Zephyr 1.6B | 53 ptsUsable (C) · 53 of 100 points: Model quality 4.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.1/15, workload match +5.0 | 92.8 tok/s | Runs well | 52 ptsUsable (C) · 52 of 100 points: Model quality 4.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.4/15, workload match +5.0 | 131 tok/s | Runs well |
| Falcon 7B Instruct | 53 ptsUsable (C) · 53 of 100 points: Model quality 2.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.8/15, workload match +5.0 | 45.8 tok/s | Runs well | 52 ptsUsable (C) · 52 of 100 points: Model quality 2.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.1/15, workload match +5.0 | 64.4 tok/s | Runs well |
| StarCoder2 3B | 53 ptsUsable (C) · 53 of 100 points: Model quality 8.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 5.2/15 | 50.3 tok/s | Runs well | 52 ptsUsable (C) · 52 of 100 points: Model quality 8.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.5/15 | 70.7 tok/s | Runs well |
| Qwen1.5 110B Chat | 53 ptsUsable (C) · 53 of 100 points: Model quality 16.4/45, Memory fit 10.0/25, Speed 6.8/15, Memory use 15.0/15, workload match +5.0 | 8.1 tok/s | Runs with offload | 53 ptsUsable (C) · 53 of 100 points: Model quality 19.7/45, Memory fit 10.0/25, Speed 4.3/15, Memory use 15.0/15, workload match +5.0 | 3.2 tok/s | Runs with offload |
| LFM2-VL 450M | 53 ptsUsable (C) · 53 of 100 points: Model quality 5.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.6/15, workload match +5.0 | 338 tok/s | Runs well | 52 ptsUsable (C) · 52 of 100 points: Model quality 5.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.9/15, workload match +5.0 | 476 tok/s | Runs well |
| LFM2 350M | 52 ptsUsable (C) · 52 of 100 points: Model quality 5.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.5/15, workload match +5.0 | 435 tok/s | Runs well | 52 ptsUsable (C) · 52 of 100 points: Model quality 5.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.8/15, workload match +5.0 | 612 tok/s | Runs well |
| Granite 4.0 H 350M | 52 ptsUsable (C) · 52 of 100 points: Model quality 5.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.5/15, workload match +5.0 | 448 tok/s | Runs well | 52 ptsUsable (C) · 52 of 100 points: Model quality 5.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.8/15, workload match +5.0 | 630 tok/s | Runs well |
| Qwen2.5-Math 1.5B Instruct | 52 ptsUsable (C) · 52 of 100 points: Model quality 9.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.7/15 | 98.9 tok/s | Runs well | 51 ptsUsable (C) · 51 of 100 points: Model quality 9.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.0/15 | 139 tok/s | Runs well |
| Devstral 2 (123B) | 52 ptsUsable (C) · 52 of 100 points: Model quality 27.1/45, Memory fit 4.0/25, Speed 6.4/15, Memory use 15.0/15 | 7.2 tok/s | Heavy offload | 57 ptsUsable (C) · 57 of 100 points: Model quality 27.1/45, Memory fit 10.0/25, Speed 5.1/15, Memory use 15.0/15 | 4.6 tok/s | Runs with offload |
| SmolVLM 500M Instruct | 52 ptsUsable (C) · 52 of 100 points: Model quality 4.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.8/15, workload match +5.0 | 299 tok/s | Runs well | 51 ptsUsable (C) · 51 of 100 points: Model quality 4.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.1/15, workload match +5.0 | 420 tok/s | Runs well |
| Qwen2.5 0.5B Instruct | 52 ptsUsable (C) · 52 of 100 points: Model quality 4.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.7/15, workload match +5.0 | 311 tok/s | Runs well | 51 ptsUsable (C) · 51 of 100 points: Model quality 4.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.0/15, workload match +5.0 | 437 tok/s | Runs well |
| LFM2.5 230M | 51 ptsUsable (C) · 51 of 100 points: Model quality 4.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.4/15, workload match +5.0 | 662 tok/s | Runs well | 51 ptsUsable (C) · 51 of 100 points: Model quality 4.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.7/15, workload match +5.0 | 932 tok/s | Runs well |
| Hunyuan 0.5B Instruct | 51 ptsUsable (C) · 51 of 100 points: Model quality 3.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.0/15, workload match +5.0 | 282 tok/s | Runs well | 50 ptsUsable (C) · 50 of 100 points: Model quality 3.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.3/15, workload match +5.0 | 397 tok/s | Runs well |
| Yi Coder 1.5B Chat | 51 ptsUsable (C) · 51 of 100 points: Model quality 7.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.3/15 | 103 tok/s | Runs well | 50 ptsUsable (C) · 50 of 100 points: Model quality 7.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.6/15 | 145 tok/s | Runs well |
| SmolLM2 360M Instruct | 51 ptsUsable (C) · 51 of 100 points: Model quality 3.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.6/15, workload match +5.0 | 423 tok/s | Runs well | 50 ptsUsable (C) · 50 of 100 points: Model quality 3.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.9/15, workload match +5.0 | 595 tok/s | Runs well |
| CodeGemma 2B | 50 ptsUsable (C) · 50 of 100 points: Model quality 6.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.6/15 | 60.7 tok/s | Runs well | 50 ptsUsable (C) · 50 of 100 points: Model quality 6.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.9/15 | 85.4 tok/s | Runs well |
| TinyLlama 1.1B Chat | 50 ptsUsable (C) · 50 of 100 points: Model quality 2.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.2/15, workload match +5.0 | 138 tok/s | Runs well | 50 ptsUsable (C) · 50 of 100 points: Model quality 2.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.5/15, workload match +5.0 | 195 tok/s | Runs well |
| Qwen2 0.5B Instruct | 50 ptsUsable (C) · 50 of 100 points: Model quality 2.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.7/15, workload match +5.0 | 311 tok/s | Runs well | 49 ptsUsable (C) · 49 of 100 points: Model quality 2.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.0/15, workload match +5.0 | 437 tok/s | Runs well |
| ERNIE 4.5 0.3B | 50 ptsUsable (C) · 50 of 100 points: Model quality 2.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.6/15, workload match +5.0 | 423 tok/s | Runs well | 49 ptsUsable (C) · 49 of 100 points: Model quality 2.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.9/15, workload match +5.0 | 595 tok/s | Runs well |
| SmolVLM 256M Instruct | 50 ptsUsable (C) · 50 of 100 points: Model quality 2.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.5/15, workload match +5.0 | 586 tok/s | Runs well | 49 ptsUsable (C) · 49 of 100 points: Model quality 2.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.8/15, workload match +5.0 | 824 tok/s | Runs well |
| Gemma 3 270M | 50 ptsUsable (C) · 50 of 100 points: Model quality 2.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.4/15, workload match +5.0 | 564 tok/s | Runs well | 49 ptsUsable (C) · 49 of 100 points: Model quality 2.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.7/15, workload match +5.0 | 794 tok/s | Runs well |
| SmolLM 360M Instruct | 49 ptsUsable (C) · 49 of 100 points: Model quality 2.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.5/15, workload match +5.0 | 423 tok/s | Runs well | 49 ptsUsable (C) · 49 of 100 points: Model quality 2.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.8/15, workload match +5.0 | 595 tok/s | Runs well |
| DeepSeek Coder 1.3B Instruct | 49 ptsUsable (C) · 49 of 100 points: Model quality 5.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.2/15 | 113 tok/s | Runs well | 48 ptsUsable (C) · 48 of 100 points: Model quality 5.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.5/15 | 159 tok/s | Runs well |
| SmolLM2 135M Instruct | 49 ptsUsable (C) · 49 of 100 points: Model quality 1.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.4/15, workload match +5.0 | 1,171 tok/s | Runs well | 48 ptsUsable (C) · 48 of 100 points: Model quality 1.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.7/15, workload match +5.0 | 1,648 tok/s | Runs well |
| Falcon-H1 Tiny 90M Instruct | 48 ptsUsable (C) · 48 of 100 points: Model quality 1.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.3/15, workload match +5.0 | 1,692 tok/s | Runs well | 47 ptsUsable (C) · 47 of 100 points: Model quality 1.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.6/15, workload match +5.0 | 2,381 tok/s | Runs well |
| SmolLM 135M Instruct | 48 ptsUsable (C) · 48 of 100 points: Model quality 0.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.3/15, workload match +5.0 | 1,171 tok/s | Runs well | 47 ptsUsable (C) · 47 of 100 points: Model quality 0.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.6/15, workload match +5.0 | 1,648 tok/s | Runs well |
| Qwen2.5 Coder 0.5B Instruct | 46 ptsUsable (C) · 46 of 100 points: Model quality 3.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.7/15 | 311 tok/s | Runs well | 45 ptsUsable (C) · 45 of 100 points: Model quality 3.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.0/15 | 437 tok/s | Runs well |
| gpt-oss 120B | 11 ptsWon't run (F) · 11 of 100 points: Model quality 11.3/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 11 ptsWon't run (F) · 11 of 100 points: Model quality 11.3/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| MiMo V2.6 Pro (1T-A42B) | 9 ptsWon't run (F) · 9 of 100 points: Model quality 9.4/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 9 ptsWon't run (F) · 9 of 100 points: Model quality 9.4/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Qwen3.8 Max (2.4T-A95B) | 9 ptsWon't run (F) · 9 of 100 points: Model quality 9.3/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 9 ptsWon't run (F) · 9 of 100 points: Model quality 9.3/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Kimi K3 (2.8T) | 9 ptsWon't run (F) · 9 of 100 points: Model quality 9.3/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 9 ptsWon't run (F) · 9 of 100 points: Model quality 9.3/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| GLM-5.3 (753B-A40B) | 9 ptsWon't run (F) · 9 of 100 points: Model quality 9.2/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 9 ptsWon't run (F) · 9 of 100 points: Model quality 9.2/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| MiMo V2.6 Flash (310B-A15B) | 9 ptsWon't run (F) · 9 of 100 points: Model quality 9.2/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 9 ptsWon't run (F) · 9 of 100 points: Model quality 9.2/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| DeepSeek V4 Pro (1.6T-A49B) | 9 ptsWon't run (F) · 9 of 100 points: Model quality 9.1/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 9 ptsWon't run (F) · 9 of 100 points: Model quality 9.1/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Qwen3.8 Flash Next (125B-A6B) | 9 ptsWon't run (F) · 9 of 100 points: Model quality 9.1/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 9 ptsWon't run (F) · 9 of 100 points: Model quality 9.1/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| GLM-5.2 (753B-A40B) | 9 ptsWon't run (F) · 9 of 100 points: Model quality 9.1/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 9 ptsWon't run (F) · 9 of 100 points: Model quality 9.1/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Hunyuan Hy4 Preview (780B) | 9 ptsWon't run (F) · 9 of 100 points: Model quality 9.1/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 9 ptsWon't run (F) · 9 of 100 points: Model quality 9.1/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Kimi K2.6 (1T) | 9 ptsWon't run (F) · 9 of 100 points: Model quality 9.0/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 9 ptsWon't run (F) · 9 of 100 points: Model quality 9.0/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| GLM-5.3 Flash (320B-A18B) | 9 ptsWon't run (F) · 9 of 100 points: Model quality 9.0/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 9 ptsWon't run (F) · 9 of 100 points: Model quality 9.0/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| GLM-5.1 (744B-A40B) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 9.0/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 9.0/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| DeepSeek V4 Pro (April 2026) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 9.0/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 9.0/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Qwen3.5 397B-A17B | 8 ptsWon't run (F) · 8 of 100 points: Model quality 9.0/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 9.0/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| MiniMax M3 (427B) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 9.0/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 9.0/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| GLM-5 (744B-A40B) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.9/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.9/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Hunyuan Hy3 (295B-A21B) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.9/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.9/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Kimi K2.5 (1T) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.9/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.9/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Kimi K2.7 Code (1T) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.9/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.9/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| DeepSeek V4 Flash (284B-A13B) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.9/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.9/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Ornith 1.5 397B-A17B | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.9/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.9/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| DeepSeek V3.2 (671B) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.9/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.9/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| GLM-4.7 (355B-A32B) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.8/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.8/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| MiniMax M2.7 (230B-A10B) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.8/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.8/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Nemotron 3 Ultra 550B-A55B | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.8/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.8/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| DeepSeek R1 0528 (671B) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.8/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.8/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Kimi K2 Thinking (1T) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.8/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.8/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| DeepSeek V3.2 Exp | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.7/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.7/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| MiniMax M2.5 (230B-A10B) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.7/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.7/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| DeepSeek R1 (671B) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.7/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.7/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| DeepSeek V4 Flash (April 2026) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.7/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.7/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Step 3.5 Flash (196B-A11B) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.7/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.7/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| DeepSeek V3.1 Terminus | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| DeepSeek V3.1 (671B) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Kimi K2 Instruct 0905 (1T) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| MiniMax M2.1 (230B-A10B) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Qwen3 235B-A22B Thinking 2507 | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Qwen3 Coder 480B-A35B | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| MiMo V2 Flash (309B-A15B) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Ling 1T | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.5/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.5/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Mistral Large 3 (675B) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.5/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.5/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Qwen3 235B-A22B Instruct 2507 | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.5/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.5/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Solar Open 2 250B | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.5/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.5/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Kimi K2 Instruct (July 2025) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.4/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.4/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| MiniMax M2 (230B-A10B) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.4/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.4/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Qwen3-VL 235B-A22B Instruct | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.4/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.4/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| K-EXAONE 236B-A23B | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.3/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.3/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| LongCat Flash Chat (560B) | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.3/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.3/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Qwen3 235B-A22B | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.3/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.3/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| InternVL3.5 241B-A28B | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.1/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.1/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Llama 3.1 Nemotron Ultra 253B v1 | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.1/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 8 ptsWon't run (F) · 8 of 100 points: Model quality 8.1/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Baichuan M3 235B-A22B | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.9/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.9/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| GLM-4.6 (355B-A32B) | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.9/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.9/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| DeepSeek V3 0324 (671B) | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.8/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.8/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| MiniMax-M1 80K | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.8/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.8/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| GLM-4.5 (355B-A32B) | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.7/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.7/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| MiniMax-M1 40K | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.7/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.7/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Trinity Large Thinking (400B-A13B) | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| ERNIE 4.5 300B-A47B | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Hermes 4 405B | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| dots.llm1 Instruct | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.2/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 56 ptsUsable (C) · 56 of 100 points: Model quality 23.9/45, Memory fit 4.0/25, Speed 8.1/15, Memory use 15.0/15, workload match +5.0 | 11.6 tok/s | Heavy offload |
| MiniMax-Text-01 | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.2/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.2/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Tülu 3 405B | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.2/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.2/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| DeepSeek V3 (671B) | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.1/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 7 ptsWon't run (F) · 7 of 100 points: Model quality 7.1/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| DeepSeek Coder V2 (236B) | 6 ptsWon't run (F) · 6 of 100 points: Model quality 7.0/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 6 ptsWon't run (F) · 6 of 100 points: Model quality 7.0/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| DeepSeek Coder V2 Instruct (0614) | 6 ptsWon't run (F) · 6 of 100 points: Model quality 6.8/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 6 ptsWon't run (F) · 6 of 100 points: Model quality 6.8/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| DeepSeek V2.5 1210 | 6 ptsWon't run (F) · 6 of 100 points: Model quality 6.8/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 6 ptsWon't run (F) · 6 of 100 points: Model quality 6.8/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Llama 3.1 405B Instruct | 6 ptsWon't run (F) · 6 of 100 points: Model quality 6.8/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 6 ptsWon't run (F) · 6 of 100 points: Model quality 6.8/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Llama 4 Maverick (17B x 128E) | 6 ptsWon't run (F) · 6 of 100 points: Model quality 6.8/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 6 ptsWon't run (F) · 6 of 100 points: Model quality 6.8/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| DeepSeek V2.5 | 6 ptsWon't run (F) · 6 of 100 points: Model quality 6.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 6 ptsWon't run (F) · 6 of 100 points: Model quality 6.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Jamba 1.5 Large | 6 ptsWon't run (F) · 6 of 100 points: Model quality 6.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 6 ptsWon't run (F) · 6 of 100 points: Model quality 6.6/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| DeepSeek V2 Chat | 5 ptsWon't run (F) · 5 of 100 points: Model quality 5.9/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 5 ptsWon't run (F) · 5 of 100 points: Model quality 5.9/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Mixtral 8x22B Instruct | 5 ptsWon't run (F) · 5 of 100 points: Model quality 5.2/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 52 ptsUsable (C) · 52 of 100 points: Model quality 17.3/45, Memory fit 10.0/25, Speed 5.7/15, Memory use 15.0/15, workload match +5.0 | 5.7 tok/s | Runs with offload |
| Grok-1 (314B) | 4 ptsWon't run (F) · 4 of 100 points: Model quality 4.7/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 4 ptsWon't run (F) · 4 of 100 points: Model quality 4.7/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Snowflake Arctic Instruct | 4 ptsWon't run (F) · 4 of 100 points: Model quality 4.7/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 4 ptsWon't run (F) · 4 of 100 points: Model quality 4.7/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
| Falcon 180B Chat | 3 ptsWon't run (F) · 3 of 100 points: Model quality 3.2/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run | 3 ptsWon't run (F) · 3 of 100 points: Model quality 3.2/45, Memory fit 0.0/25, Speed 0.0/15, Memory use 0.0/15 | — | Won't run |
Image models
| Model | Grade · points (Apple M5 Max (64 GB)) | Time (Apple M5 Max (64 GB)) | Fit (Apple M5 Max (64 GB)) | Grade · points (AMD Radeon PRO W7800 48GB) | Time (AMD Radeon PRO W7800 48GB) | Fit (AMD Radeon PRO W7800 48GB) |
|---|---|---|---|---|---|---|
| FLUX.2 [klein] 9B | 92 ptsExcellent (S) · 92 of 100 points: Model quality 37.8/45, Memory fit 25.0/25, Speed 14.8/15, Memory use 15.0/15 | 10 s | Runs well | 92 ptsExcellent (S) · 92 of 100 points: Model quality 37.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | 2.5 s | Runs well |
| Z-Image Turbo | 86 ptsExcellent (S) · 86 of 100 points: Model quality 37.8/45, Memory fit 25.0/25, Speed 12.5/15, Memory use 10.9/15 | 14 s | Runs well | 88 ptsExcellent (S) · 88 of 100 points: Model quality 37.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 10.9/15 | 3.5 s | Runs well |
| FLUX.2 [klein] 4B | 83 ptsGreat (A) · 83 of 100 points: Model quality 35.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.7/15 | 5.7 s | Runs well | 83 ptsGreat (A) · 83 of 100 points: Model quality 35.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.7/15 | 1.4 s | Runs well |
| FLUX.1 Krea [dev] | 82 ptsGreat (A) · 82 of 100 points: Model quality 36.9/45, Memory fit 25.0/25, Speed 5.7/15, Memory use 15.0/15 | 69 s | Runs well | 88 ptsExcellent (S) · 88 of 100 points: Model quality 36.9/45, Memory fit 25.0/25, Speed 11.6/15, Memory use 15.0/15 | 17 s | Runs well |
| Stable Diffusion 3.5 Large Turbo | 82 ptsGreat (A) · 82 of 100 points: Model quality 27.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 14.4/15 | 9.2 s | Runs well | 82 ptsGreat (A) · 82 of 100 points: Model quality 27.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 14.4/15 | 2.2 s | Runs well |
| Qwen-Image 2512 | 82 ptsGreat (A) · 82 of 100 points: Model quality 40.1/45, Memory fit 25.0/25, Speed 1.9/15, Memory use 15.0/15 | 9.9 min | Runs well | 84 ptsGreat (A) · 84 of 100 points: Model quality 39.7/45, Memory fit 25.0/25, Speed 4.5/15, Memory use 15.0/15 | 1.9 min | Runs well |
| FLUX.1 [dev] | 81 ptsGreat (A) · 81 of 100 points: Model quality 36.0/45, Memory fit 25.0/25, Speed 5.7/15, Memory use 15.0/15 | 69 s | Runs well | 87 ptsExcellent (S) · 87 of 100 points: Model quality 36.0/45, Memory fit 25.0/25, Speed 11.6/15, Memory use 15.0/15 | 17 s | Runs well |
| FLUX.1 Kontext [dev] | 81 ptsGreat (A) · 81 of 100 points: Model quality 36.9/45, Memory fit 25.0/25, Speed 4.2/15, Memory use 15.0/15 | 2.1 min | Runs well | 85 ptsExcellent (S) · 85 of 100 points: Model quality 36.9/45, Memory fit 25.0/25, Speed 8.5/15, Memory use 15.0/15 | 31 s | Runs well |
| Qwen-Image | 81 ptsGreat (A) · 81 of 100 points: Model quality 38.8/45, Memory fit 25.0/25, Speed 2.2/15, Memory use 15.0/15 | 7.8 min | Runs well | 83 ptsGreat (A) · 83 of 100 points: Model quality 38.8/45, Memory fit 25.0/25, Speed 4.5/15, Memory use 15.0/15 | 1.9 min | Runs well |
| Qwen-Image-Edit 2511 | 80 ptsGreat (A) · 80 of 100 points: Model quality 38.3/45, Memory fit 25.0/25, Speed 2.1/15, Memory use 15.0/15 | 8.8 min | Runs well | 82 ptsGreat (A) · 82 of 100 points: Model quality 37.9/45, Memory fit 25.0/25, Speed 4.7/15, Memory use 15.0/15 | 1.7 min | Runs well |
| HiDream-I1 Dev | 80 ptsGreat (A) · 80 of 100 points: Model quality 35.3/45, Memory fit 25.0/25, Speed 5.1/15, Memory use 15.0/15 | 88 s | Runs well | 85 ptsExcellent (S) · 85 of 100 points: Model quality 35.3/45, Memory fit 25.0/25, Speed 10.2/15, Memory use 15.0/15 | 21 s | Runs well |
| HiDream-I1 Full | 80 ptsGreat (A) · 80 of 100 points: Model quality 37.4/45, Memory fit 25.0/25, Speed 2.8/15, Memory use 15.0/15 | 5.0 min | Runs well | 83 ptsGreat (A) · 83 of 100 points: Model quality 37.0/45, Memory fit 25.0/25, Speed 6.3/15, Memory use 15.0/15 | 57 s | Runs well |
| HiDream-I1 Fast | 80 ptsGreat (A) · 80 of 100 points: Model quality 33.5/45, Memory fit 25.0/25, Speed 6.6/15, Memory use 15.0/15 | 52 s | Runs well | 86 ptsExcellent (S) · 86 of 100 points: Model quality 33.5/45, Memory fit 25.0/25, Speed 13.4/15, Memory use 15.0/15 | 13 s | Runs well |
| FLUX.1 [schnell] | 80 ptsGreat (A) · 80 of 100 points: Model quality 30.9/45, Memory fit 25.0/25, Speed 14.8/15, Memory use 9.4/15 | 10 s | Runs well | 86 ptsExcellent (S) · 86 of 100 points: Model quality 31.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | 2.8 s | Runs well |
| FLUX.1 Fill [dev] | 79 ptsGreat (A) · 79 of 100 points: Model quality 34.2/45, Memory fit 25.0/25, Speed 5.6/15, Memory use 15.0/15 | 72 s | Runs well | 85 ptsExcellent (S) · 85 of 100 points: Model quality 34.2/45, Memory fit 25.0/25, Speed 11.3/15, Memory use 15.0/15 | 18 s | Runs well |
| FLUX.2 [dev] | 78 ptsGreat (A) · 78 of 100 points: Model quality 41.0/45, Memory fit 20.0/25, Speed 3.0/15, Memory use 15.0/15 | 4.2 min | Runs well | 82 ptsGreat (A) · 82 of 100 points: Model quality 40.6/45, Memory fit 20.0/25, Speed 6.9/15, Memory use 15.0/15 | 48 s | Runs well |
| Chroma1-HD | 78 ptsGreat (A) · 78 of 100 points: Model quality 35.1/45, Memory fit 25.0/25, Speed 4.4/15, Memory use 14.4/15 | 1.9 min | Runs well | 83 ptsGreat (A) · 83 of 100 points: Model quality 35.1/45, Memory fit 25.0/25, Speed 9.0/15, Memory use 14.4/15 | 28 s | Runs well |
| Z-Image | 77 ptsGreat (A) · 77 of 100 points: Model quality 36.9/45, Memory fit 25.0/25, Speed 4.8/15, Memory use 10.9/15 | 1.6 min | Runs well | 82 ptsGreat (A) · 82 of 100 points: Model quality 36.9/45, Memory fit 25.0/25, Speed 9.7/15, Memory use 10.9/15 | 24 s | Runs well |
| CogView4 6B | 76 ptsGreat (A) · 76 of 100 points: Model quality 31.5/45, Memory fit 25.0/25, Speed 4.7/15, Memory use 15.0/15 | 1.7 min | Runs well | 80 ptsGreat (A) · 80 of 100 points: Model quality 31.5/45, Memory fit 25.0/25, Speed 9.4/15, Memory use 15.0/15 | 25 s | Runs well |
| Stable Diffusion 3.5 Large | 75 ptsGreat (A) · 75 of 100 points: Model quality 29.7/45, Memory fit 25.0/25, Speed 6.0/15, Memory use 14.4/15 | 63 s | Runs well | 74 ptsGreat (A) · 74 of 100 points: Model quality 29.1/45, Memory fit 25.0/25, Speed 12.7/15, Memory use 7.8/15 | 14 s | Runs well |
| Kolors | 71 ptsGood (B) · 71 of 100 points: Model quality 27.0/45, Memory fit 25.0/25, Speed 9.9/15, Memory use 9.5/15 | 23 s | Runs well | 76 ptsGreat (A) · 76 of 100 points: Model quality 27.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 9.5/15 | 5.6 s | Runs well |
| DreamShaper XL v2 Turbo | 70 ptsGood (B) · 70 of 100 points: Model quality 26.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.3/15 | 7.5 s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 26.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.3/15 | 1.8 s | Runs well |
| HunyuanImage 3.0 (80B-A13B) | 69 ptsGood (B) · 69 of 100 points: Model quality 39.7/45, Memory fit 12.0/25, Speed 2.4/15, Memory use 15.0/15 | 6.7 min | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 39.7/45, Memory fit 12.0/25, Speed 4.8/15, Memory use 15.0/15 | 1.6 min | Runs well |
| PixArt-Σ XL 1024 | 68 ptsGood (B) · 68 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15 | 8.6 s | Runs well | 68 ptsGood (B) · 68 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.2/15 | 2.1 s | Runs well |
| SDXL-Lightning (4-step) | 68 ptsGood (B) · 68 of 100 points: Model quality 24.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.3/15 | 4.0 s | Runs well | 68 ptsGood (B) · 68 of 100 points: Model quality 24.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.3/15 | 1.0 s | Runs well |
| OmniGen2 | 68 ptsGood (B) · 68 of 100 points: Model quality 29.7/45, Memory fit 25.0/25, Speed 4.9/15, Memory use 8.7/15 | 1.5 min | Runs well | 73 ptsGreat (A) · 73 of 100 points: Model quality 29.7/45, Memory fit 25.0/25, Speed 10.0/15, Memory use 8.7/15 | 22 s | Runs well |
| Stable Diffusion 3.5 Medium | 68 ptsGood (B) · 68 of 100 points: Model quality 26.1/45, Memory fit 25.0/25, Speed 8.1/15, Memory use 8.9/15 | 34 s | Runs well | 74 ptsGreat (A) · 74 of 100 points: Model quality 26.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.9/15 | 8.4 s | Runs well |
| Juggernaut XL v9 | 66 ptsGood (B) · 66 of 100 points: Model quality 27.9/45, Memory fit 25.0/25, Speed 9.3/15, Memory use 4.3/15 | 26 s | Runs well | 72 ptsGreat (A) · 72 of 100 points: Model quality 27.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.3/15 | 6.3 s | Runs well |
| RealVisXL V5.0 | 66 ptsGood (B) · 66 of 100 points: Model quality 27.9/45, Memory fit 25.0/25, Speed 9.3/15, Memory use 4.3/15 | 26 s | Runs well | 72 ptsGreat (A) · 72 of 100 points: Model quality 27.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.3/15 | 6.3 s | Runs well |
| Lumina-Image 2.0 | 66 ptsGood (B) · 66 of 100 points: Model quality 27.9/45, Memory fit 25.0/25, Speed 7.5/15, Memory use 6.1/15 | 40 s | Runs well | 74 ptsGreat (A) · 74 of 100 points: Model quality 27.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.1/15 | 9.8 s | Runs well |
| Animagine XL 4.0 | 65 ptsGood (B) · 65 of 100 points: Model quality 27.0/45, Memory fit 25.0/25, Speed 9.7/15, Memory use 4.3/15 | 24 s | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 27.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.3/15 | 5.9 s | Runs well |
| Illustrious XL v0.1 | 65 ptsGood (B) · 65 of 100 points: Model quality 27.0/45, Memory fit 25.0/25, Speed 9.7/15, Memory use 4.3/15 | 24 s | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 27.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.3/15 | 5.9 s | Runs well |
| Kandinsky 3 | 65 ptsGood (B) · 65 of 100 points: Model quality 21.6/45, Memory fit 25.0/25, Speed 6.6/15, Memory use 12.4/15 | 52 s | Runs well | 72 ptsGreat (A) · 72 of 100 points: Model quality 21.6/45, Memory fit 25.0/25, Speed 13.4/15, Memory use 12.4/15 | 13 s | Runs well |
| AuraFlow v0.3 | 65 ptsGood (B) · 65 of 100 points: Model quality 26.1/45, Memory fit 25.0/25, Speed 5.3/15, Memory use 8.9/15 | 80 s | Runs well | 66 ptsGood (B) · 66 of 100 points: Model quality 25.6/45, Memory fit 25.0/25, Speed 11.3/15, Memory use 5.0/15 | 18 s | Runs well |
| Stable Diffusion 3 Medium | 64 ptsGood (B) · 64 of 100 points: Model quality 22.9/45, Memory fit 25.0/25, Speed 11.8/15, Memory use 4.7/15 | 16 s | Runs well | 71 ptsGood (B) · 71 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.4/15 | 4.3 s | Runs well |
| Pony Diffusion V6 XL | 63 ptsGood (B) · 63 of 100 points: Model quality 25.6/45, Memory fit 25.0/25, Speed 10.7/15, Memory use 2.7/15 | 20 s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 26.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.3/15 | 5.3 s | Runs well |
| Playground v2.5 | 63 ptsGood (B) · 63 of 100 points: Model quality 27.0/45, Memory fit 25.0/25, Speed 7.2/15, Memory use 4.3/15 | 43 s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 27.0/45, Memory fit 25.0/25, Speed 14.6/15, Memory use 4.3/15 | 11 s | Runs well |
| Animagine XL 3.1 | 62 ptsGood (B) · 62 of 100 points: Model quality 24.7/45, Memory fit 25.0/25, Speed 10.2/15, Memory use 2.7/15 | 22 s | Runs well | 69 ptsGood (B) · 69 of 100 points: Model quality 25.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.3/15 | 5.9 s | Runs well |
| Stable Diffusion XL 1.0 | 61 ptsGood (B) · 61 of 100 points: Model quality 24.3/45, Memory fit 25.0/25, Speed 9.8/15, Memory use 2.7/15 | 23 s | Runs well | 69 ptsGood (B) · 69 of 100 points: Model quality 24.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.3/15 | 6.3 s | Runs well |
| SDXL Turbo | 61 ptsGood (B) · 61 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.7/15 | 0.9 s | Runs well | 61 ptsGood (B) · 61 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.7/15 | 0.2 s | Runs well |
| Stable Cascade | 61 ptsGood (B) · 61 of 100 points: Model quality 22.9/45, Memory fit 25.0/25, Speed 9.3/15, Memory use 3.9/15 | 26 s | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 6.7/15 | 7.0 s | Runs well |
| Hunyuan-DiT v1.2 | 60 ptsGood (B) · 60 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 8.1/15, Memory use 4.5/15 | 34 s | Runs well | 67 ptsGood (B) · 67 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 4.5/15 | 8.4 s | Runs well |
| Realistic Vision V5.1 | 60 ptsGood (B) · 60 of 100 points: Model quality 18.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.3/15 | 6.9 s | Runs well | 60 ptsGood (B) · 60 of 100 points: Model quality 18.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.3/15 | 1.7 s | Runs well |
| DreamShaper 8 | 59 ptsGood (B) · 59 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.3/15 | 6.9 s | Runs well | 59 ptsGood (B) · 59 of 100 points: Model quality 18.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.3/15 | 1.7 s | Runs well |
| Stable Diffusion 1.5 | 57 ptsUsable (C) · 57 of 100 points: Model quality 15.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.3/15 | 6.9 s | Runs well | 57 ptsUsable (C) · 57 of 100 points: Model quality 15.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.3/15 | 1.7 s | Runs well |
| LCM DreamShaper v7 | 54 ptsUsable (C) · 54 of 100 points: Model quality 13.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.3/15 | 1.4 s | Runs well | 54 ptsUsable (C) · 54 of 100 points: Model quality 13.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.3/15 | 0.4 s | Runs well |
| SD Turbo | 54 ptsUsable (C) · 54 of 100 points: Model quality 12.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.6/15 | 0.5 s | Runs well | 54 ptsUsable (C) · 54 of 100 points: Model quality 12.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.6/15 | 0.1 s | Runs well |
Video models
| Model | Grade · points (Apple M5 Max (64 GB)) | Time (Apple M5 Max (64 GB)) | Fit (Apple M5 Max (64 GB)) | Grade · points (AMD Radeon PRO W7800 48GB) | Time (AMD Radeon PRO W7800 48GB) | Fit (AMD Radeon PRO W7800 48GB) |
|---|---|---|---|---|---|---|
| LTX-Video 13B 0.9.8 (distilled) | 80 ptsGreat (A) · 80 of 100 points: Model quality 30.0/45, Memory fit 25.0/25, Speed 11.4/15, Memory use 14.5/15 | 5.2 min | Runs well | 84 ptsGreat (A) · 84 of 100 points: Model quality 30.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 14.5/15 | 76 s | Runs well |
| LTX-2.3 | 80 ptsGreat (A) · 80 of 100 points: Model quality 37.9/45, Memory fit 23.0/25, Speed 4.4/15, Memory use 15.0/15 | 34.5 min | Tight fit | 84 ptsGreat (A) · 84 of 100 points: Model quality 37.9/45, Memory fit 23.0/25, Speed 9.0/15, Memory use 15.0/15 | 8.4 min | Tight fit |
| Wan 2.2 I2V A14B | 80 ptsGreat (A) · 80 of 100 points: Model quality 38.3/45, Memory fit 25.0/25, Speed 2.0/15, Memory use 15.0/15 | 2.9 h | Runs well | 82 ptsGreat (A) · 82 of 100 points: Model quality 37.9/45, Memory fit 25.0/25, Speed 4.5/15, Memory use 15.0/15 | 33.6 min | Runs well |
| Wan 2.2 T2V A14B | 80 ptsGreat (A) · 80 of 100 points: Model quality 38.3/45, Memory fit 25.0/25, Speed 2.0/15, Memory use 15.0/15 | 2.9 h | Runs well | 82 ptsGreat (A) · 82 of 100 points: Model quality 37.9/45, Memory fit 25.0/25, Speed 4.5/15, Memory use 15.0/15 | 33.6 min | Runs well |
| HunyuanVideo 1.5 (8.3B) | 79 ptsGreat (A) · 79 of 100 points: Model quality 37.0/45, Memory fit 25.0/25, Speed 2.1/15, Memory use 15.0/15 | 2.6 h | Runs well | 81 ptsGreat (A) · 81 of 100 points: Model quality 37.0/45, Memory fit 25.0/25, Speed 4.2/15, Memory use 15.0/15 | 37.8 min | Runs well |
| LTX-2 (19B, video + audio) | 78 ptsGreat (A) · 78 of 100 points: Model quality 36.2/45, Memory fit 23.0/25, Speed 4.4/15, Memory use 15.0/15 | 34.5 min | Tight fit | 83 ptsGreat (A) · 83 of 100 points: Model quality 36.2/45, Memory fit 23.0/25, Speed 9.0/15, Memory use 15.0/15 | 8.4 min | Tight fit |
| Wan 2.1 I2V 14B 720P | 76 ptsGreat (A) · 76 of 100 points: Model quality 34.7/45, Memory fit 25.0/25, Speed 1.8/15, Memory use 15.0/15 | 3.7 h | Runs well | 78 ptsGreat (A) · 78 of 100 points: Model quality 34.4/45, Memory fit 25.0/25, Speed 4.0/15, Memory use 15.0/15 | 42.1 min | Runs well |
| Wan 2.1 T2V 14B | 76 ptsGreat (A) · 76 of 100 points: Model quality 34.7/45, Memory fit 25.0/25, Speed 1.8/15, Memory use 15.0/15 | 3.7 h | Runs well | 78 ptsGreat (A) · 78 of 100 points: Model quality 34.4/45, Memory fit 25.0/25, Speed 4.0/15, Memory use 15.0/15 | 42.1 min | Runs well |
| FramePack I2V (HunyuanVideo 13B) | 75 ptsGreat (A) · 75 of 100 points: Model quality 32.6/45, Memory fit 25.0/25, Speed 4.4/15, Memory use 13.5/15 | 34.5 min | Runs well | 80 ptsGreat (A) · 80 of 100 points: Model quality 32.6/45, Memory fit 25.0/25, Speed 9.0/15, Memory use 13.5/15 | 8.4 min | Runs well |
| Wan 2.2 S2V 14B (speech to video) | 75 ptsGreat (A) · 75 of 100 points: Model quality 33.9/45, Memory fit 25.0/25, Speed 2.5/15, Memory use 14.1/15 | 1.8 h | Runs well | 77 ptsGreat (A) · 77 of 100 points: Model quality 33.5/45, Memory fit 25.0/25, Speed 5.7/15, Memory use 13.5/15 | 21.0 min | Runs well |
| Wan 2.2 TI2V 5B | 74 ptsGreat (A) · 74 of 100 points: Model quality 30.9/45, Memory fit 25.0/25, Speed 3.9/15, Memory use 15.0/15 | 44.8 min | Runs well | 78 ptsGreat (A) · 78 of 100 points: Model quality 30.9/45, Memory fit 25.0/25, Speed 7.9/15, Memory use 15.0/15 | 10.9 min | Runs well |
| Wan 2.2 Animate 14B | 74 ptsGreat (A) · 74 of 100 points: Model quality 34.7/45, Memory fit 23.0/25, Speed 1.8/15, Memory use 15.0/15 | 3.5 h | Tight fit | 76 ptsGreat (A) · 76 of 100 points: Model quality 34.4/45, Memory fit 23.0/25, Speed 4.1/15, Memory use 15.0/15 | 40.0 min | Tight fit |
| Wan 2.1 VACE 14B | 74 ptsGreat (A) · 74 of 100 points: Model quality 34.7/45, Memory fit 23.0/25, Speed 1.7/15, Memory use 15.0/15 | 4.0 h | Tight fit | 76 ptsGreat (A) · 76 of 100 points: Model quality 34.4/45, Memory fit 23.0/25, Speed 3.8/15, Memory use 15.0/15 | 46.3 min | Tight fit |
| SkyReels V2 T2V 14B 720P | 74 ptsGreat (A) · 74 of 100 points: Model quality 34.4/45, Memory fit 23.0/25, Speed 1.8/15, Memory use 15.0/15 | 3.4 h | Tight fit | 76 ptsGreat (A) · 76 of 100 points: Model quality 34.4/45, Memory fit 23.0/25, Speed 3.7/15, Memory use 15.0/15 | 50.5 min | Tight fit |
| SkyReels V2 T2V 14B 540P | 74 ptsGreat (A) · 74 of 100 points: Model quality 32.6/45, Memory fit 25.0/25, Speed 2.4/15, Memory use 14.1/15 | 2.0 h | Runs well | 76 ptsGreat (A) · 76 of 100 points: Model quality 32.6/45, Memory fit 25.0/25, Speed 4.8/15, Memory use 14.1/15 | 29.4 min | Runs well |
| Wan 2.1 I2V 14B 480P | 73 ptsGreat (A) · 73 of 100 points: Model quality 32.6/45, Memory fit 25.0/25, Speed 3.0/15, Memory use 12.6/15 | 77.5 min | Runs well | 76 ptsGreat (A) · 76 of 100 points: Model quality 32.6/45, Memory fit 25.0/25, Speed 6.0/15, Memory use 12.6/15 | 18.9 min | Runs well |
| Cosmos-Predict2 14B Video2World | 73 ptsGreat (A) · 73 of 100 points: Model quality 30.9/45, Memory fit 25.0/25, Speed 2.2/15, Memory use 15.0/15 | 2.3 h | Runs well | 75 ptsGreat (A) · 75 of 100 points: Model quality 30.9/45, Memory fit 25.0/25, Speed 4.5/15, Memory use 15.0/15 | 33.6 min | Runs well |
| HunyuanVideo 13B | 71 ptsGood (B) · 71 of 100 points: Model quality 30.1/45, Memory fit 25.0/25, Speed 1.6/15, Memory use 15.0/15 | 4.6 h | Runs well | 73 ptsGreat (A) · 73 of 100 points: Model quality 31.8/45, Memory fit 23.0/25, Speed 3.7/15, Memory use 15.0/15 | 50.5 min | Tight fit |
| HunyuanVideo I2V 13B | 71 ptsGood (B) · 71 of 100 points: Model quality 30.1/45, Memory fit 25.0/25, Speed 1.6/15, Memory use 15.0/15 | 4.6 h | Runs well | 73 ptsGreat (A) · 73 of 100 points: Model quality 31.8/45, Memory fit 23.0/25, Speed 3.7/15, Memory use 15.0/15 | 50.5 min | Tight fit |
| Cosmos-Predict2 2B Video2World | 71 ptsGood (B) · 71 of 100 points: Model quality 26.1/45, Memory fit 25.0/25, Speed 5.3/15, Memory use 15.0/15 | 23.9 min | Runs well | 74 ptsGreat (A) · 74 of 100 points: Model quality 25.6/45, Memory fit 25.0/25, Speed 11.3/15, Memory use 12.9/15 | 5.3 min | Runs well |
| Open-Sora 2.0 (11B) | 71 ptsGood (B) · 71 of 100 points: Model quality 31.5/45, Memory fit 23.0/25, Speed 1.5/15, Memory use 15.0/15 | 4.8 h | Tight fit | 72 ptsGreat (A) · 72 of 100 points: Model quality 31.5/45, Memory fit 23.0/25, Speed 3.1/15, Memory use 15.0/15 | 70.1 min | Tight fit |
| LTX-Video 2B | 70 ptsGood (B) · 70 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.4/15 | 2.4 min | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 22.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.4/15 | 35 s | Runs well |
| CogVideoX 1.5 5B | 69 ptsGood (B) · 69 of 100 points: Model quality 26.5/45, Memory fit 25.0/25, Speed 3.0/15, Memory use 15.0/15 | 77.5 min | Runs well | 72 ptsGreat (A) · 72 of 100 points: Model quality 26.5/45, Memory fit 25.0/25, Speed 6.0/15, Memory use 15.0/15 | 18.9 min | Runs well |
| CogVideoX 5B I2V | 68 ptsGood (B) · 68 of 100 points: Model quality 24.8/45, Memory fit 25.0/25, Speed 6.3/15, Memory use 12.2/15 | 17.2 min | Runs well | 69 ptsGood (B) · 69 of 100 points: Model quality 24.3/45, Memory fit 25.0/25, Speed 13.4/15, Memory use 7.2/15 | 3.8 min | Runs well |
| Step-Video T2V (30B) | 68 ptsGood (B) · 68 of 100 points: Model quality 33.5/45, Memory fit 18.0/25, Speed 1.6/15, Memory use 15.0/15 | 4.7 h | Tight fit | 69 ptsGood (B) · 69 of 100 points: Model quality 33.5/45, Memory fit 18.0/25, Speed 3.1/15, Memory use 15.0/15 | 68.4 min | Tight fit |
| Mochi 1 Preview (10B) | 67 ptsGood (B) · 67 of 100 points: Model quality 27.3/45, Memory fit 25.0/25, Speed 2.7/15, Memory use 12.8/15 | 1.6 h | Runs well | 70 ptsGood (B) · 70 of 100 points: Model quality 27.3/45, Memory fit 25.0/25, Speed 5.4/15, Memory use 12.8/15 | 23.1 min | Runs well |
| CogVideoX 5B | 67 ptsGood (B) · 67 of 100 points: Model quality 24.8/45, Memory fit 25.0/25, Speed 6.3/15, Memory use 11.6/15 | 17.2 min | Runs well | 69 ptsGood (B) · 69 of 100 points: Model quality 24.3/45, Memory fit 25.0/25, Speed 13.4/15, Memory use 6.9/15 | 3.8 min | Runs well |
| Allegro (2.8B) | 65 ptsGood (B) · 65 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 2.4/15, Memory use 15.0/15 | 1.9 h | Runs well | 68 ptsGood (B) · 68 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 4.9/15, Memory use 15.0/15 | 28.0 min | Runs well |
| Cosmos 1.0 Diffusion 7B Text2World | 65 ptsGood (B) · 65 of 100 points: Model quality 27.0/45, Memory fit 20.0/25, Speed 3.5/15, Memory use 15.0/15 | 55.3 min | Runs well | 69 ptsGood (B) · 69 of 100 points: Model quality 27.0/45, Memory fit 20.0/25, Speed 7.1/15, Memory use 15.0/15 | 13.5 min | Runs well |
| Wan 2.1 VACE 1.3B | 65 ptsGood (B) · 65 of 100 points: Model quality 24.8/45, Memory fit 25.0/25, Speed 5.2/15, Memory use 10.4/15 | 24.9 min | Runs well | 66 ptsGood (B) · 66 of 100 points: Model quality 24.3/45, Memory fit 25.0/25, Speed 11.1/15, Memory use 6.6/15 | 5.5 min | Runs well |
| Wan 2.1 T2V 1.3B | 64 ptsGood (B) · 64 of 100 points: Model quality 24.8/45, Memory fit 25.0/25, Speed 5.4/15, Memory use 9.5/15 | 23.0 min | Runs well | 66 ptsGood (B) · 66 of 100 points: Model quality 24.3/45, Memory fit 25.0/25, Speed 11.6/15, Memory use 6.2/15 | 5.0 min | Runs well |
| SkyReels V2 Diffusion Forcing 1.3B | 64 ptsGood (B) · 64 of 100 points: Model quality 23.4/45, Memory fit 25.0/25, Speed 4.8/15, Memory use 11.5/15 | 28.7 min | Runs well | 66 ptsGood (B) · 66 of 100 points: Model quality 22.9/45, Memory fit 25.0/25, Speed 10.3/15, Memory use 8.0/15 | 6.3 min | Runs well |
| CogVideoX 2B | 61 ptsGood (B) · 61 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 11.4/15, Memory use 5.3/15 | 5.2 min | Runs well | 68 ptsGood (B) · 68 of 100 points: Model quality 20.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 8.4/15 | 84 s | Runs well |
| AnimateDiff v3 (SD 1.5) | 58 ptsGood (B) · 58 of 100 points: Model quality 15.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.3/15 | 1.9 min | Runs well | 58 ptsGood (B) · 58 of 100 points: Model quality 15.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.3/15 | 28 s | Runs well |
Swap sides All models on the Apple M5 Max (64 GB) All models on the AMD Radeon PRO W7800 48GB
Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.