Hardware · AMD
What AI models can the AMD Instinct MI350X run?
- VRAM
- 288 GB
- Usable for models
- 288 GB
- Memory bandwidth
- 8,000 GB/s
- FP16 compute
- 137.0 TFLOPS
- Released
- 2025
Specifications
| Maker | AMD |
|---|---|
| Type | Datacenter GPU module |
| Video memory (VRAM) | 288 GB |
| Usable for AI models | 288 GB |
| Memory bandwidth | 8,000 GB/s |
| FP16 compute | 137.0 TFLOPS |
| Interface | PCIe 5.0 |
| Released | 2025 |
Best LLMs for the AMD Instinct MI350X
| Grade · points | Model | Quant | Memory | Speed | Fit |
|---|---|---|---|---|---|
| 98 ptsExcellent (S) · 98 of 100 points: Model quality 43.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | Qwen3.8 Flash Next (125B-A6B) | Q8_0 | 206 GB | 359 tok/s | Runs well |
| 97 ptsExcellent (S) · 97 of 100 points: Model quality 42.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | MiMo V2.6 Flash (310B-A15B) | Q5_K_M | 236 GB | 233 tok/s | Runs well |
| 96 ptsExcellent (S) · 96 of 100 points: Model quality 41.7/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | MiniMax M2.7 (230B-A10B) | Q6_K | 205 GB | 293 tok/s | Runs well |
| 96 ptsExcellent (S) · 96 of 100 points: Model quality 41.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | GLM-5.3 Flash (320B-A18B) | Q5_K_M | 243 GB | 194 tok/s | Runs well |
| 96 ptsExcellent (S) · 96 of 100 points: Model quality 41.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | Step 3.5 Flash (196B-A11B) | Q8_0 | 227 GB | 196 tok/s | Runs well |
| 96 ptsExcellent (S) · 96 of 100 points: Model quality 41.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | MiniMax M2.5 (230B-A10B) | Q6_K | 205 GB | 293 tok/s | Runs well |
| 96 ptsExcellent (S) · 96 of 100 points: Model quality 41.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | Hunyuan Hy3 (295B-A21B) | Q5_K_M | 230 GB | 166 tok/s | Runs well |
| 96 ptsExcellent (S) · 96 of 100 points: Model quality 41.0/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | DeepSeek V4 Flash (284B-A13B) | Q5_K_M | 231 GB | 269 tok/s | Runs well |
| 95 ptsExcellent (S) · 95 of 100 points: Model quality 40.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | MiniMax M2.1 (230B-A10B) | Q6_K | 205 GB | 293 tok/s | Runs well |
| 95 ptsExcellent (S) · 95 of 100 points: Model quality 40.5/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | Qwen3 235B-A22B Thinking 2507 | Q6_K | 209 GB | 133 tok/s | Runs well |
| 95 ptsExcellent (S) · 95 of 100 points: Model quality 40.2/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | DeepSeek V4 Flash (April 2026) | Q5_K_M | 222 GB | 269 tok/s | Runs well |
| 95 ptsExcellent (S) · 95 of 100 points: Model quality 40.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | Qwen3 235B-A22B Instruct 2507 | Q6_K | 209 GB | 133 tok/s | Runs well |
Best Image models for the AMD Instinct MI350X
| Grade · points | Model | Precision | Memory | Time per output | Fit |
|---|---|---|---|---|---|
| 85 ptsExcellent (S) · 85 of 100 points: Model quality 40.5/45, Memory fit 25.0/25, Speed 6.0/15, Memory use 13.6/15 | HunyuanImage 3.0 (80B-A13B) | FP16 | 170 GB | 62 s | Runs well |
| 83 ptsGreat (A) · 83 of 100 points: Model quality 41.4/45, Memory fit 25.0/25, Speed 7.8/15, Memory use 9.3/15 | FLUX.2 [dev] | FP16 | 116 GB | 37 s | Runs well |
| 80 ptsGreat (A) · 80 of 100 points: Model quality 37.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.0/15 | FLUX.2 [klein] 9B | FP16 | 37.1 GB | 1.8 s | Runs well |
| 79 ptsGreat (A) · 79 of 100 points: Model quality 37.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 1.8/15 | Z-Image Turbo | FP16 | 22.7 GB | 2.6 s | Runs well |
| 79 ptsGreat (A) · 79 of 100 points: Model quality 34.2/45, Memory fit 25.0/25, Speed 14.8/15, Memory use 5.2/15 | HiDream-I1 Fast | FP16 | 64.4 GB | 10 s | Runs well |
| 78 ptsGreat (A) · 78 of 100 points: Model quality 36.9/45, Memory fit 25.0/25, Speed 13.5/15, Memory use 2.9/15 | FLUX.1 Krea [dev] | FP16 | 36.4 GB | 12 s | Runs well |
Best Video models for the AMD Instinct MI350X
| Grade · points | Model | Precision | Memory | Time per output | Fit |
|---|---|---|---|---|---|
| 79 ptsGreat (A) · 79 of 100 points: Model quality 38.7/45, Memory fit 25.0/25, Speed 9.9/15, Memory use 6.0/15 | LTX-2.3 | FP16 | 75.4 GB | 6.8 min | Runs well |
| 77 ptsGreat (A) · 77 of 100 points: Model quality 36.9/45, Memory fit 25.0/25, Speed 9.9/15, Memory use 6.0/15 | LTX-2 (19B, video + audio) | FP16 | 75.4 GB | 6.8 min | Runs well |
| 74 ptsGreat (A) · 74 of 100 points: Model quality 30.6/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 3.8/15 | LTX-Video 13B 0.9.8 (distilled) | FP16 | 47.9 GB | 62 s | Runs well |
| 73 ptsGreat (A) · 73 of 100 points: Model quality 38.7/45, Memory fit 25.0/25, Speed 5.0/15, Memory use 4.6/15 | Wan 2.2 I2V A14B | FP16 | 58.0 GB | 27.4 min | Runs well |
| 73 ptsGreat (A) · 73 of 100 points: Model quality 38.7/45, Memory fit 25.0/25, Speed 5.0/15, Memory use 4.6/15 | Wan 2.2 T2V A14B | FP16 | 58.0 GB | 27.4 min | Runs well |
Best LLMs by task on this hardware Open in Model Finder
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.