Tier list
Local AI model tier list
Every model we track, ranked by quality alone. Tiers don’t depend on your hardware: they tell you how capable a model is. To see which ones your machine can run, use the Model Finder.
Step 1
Tell us your computer first
Then you will see how many of these 536 LLMs run on it, and can filter the list by how well each one runs.
S
Excellent · 85+ of 100
Show or hide the list
- MiMo V2.6 Pro (1T-A42B)
- Qwen3.8 Max (2.4T-A95B)
- GLM-5.3 (753B-A40B)
- MiMo V2.6 Flash (310B-A15B)
- DeepSeek V4 Pro (1.6T-A49B)
- Qwen3.8 Flash Next (125B-A6B)
- GLM-5.2 (753B-A40B)
- Hunyuan Hy4 Preview (780B)
- Kimi K2.6 (1T)
- GLM-5.3 Flash (320B-A18B)
- GLM-5.1 (744B-A40B)
- DeepSeek V4 Pro (April 2026)
- Qwen3.5 397B-A17B
- MiniMax M3 (427B)
- GLM-5 (744B-A40B)
- Hunyuan Hy3 (295B-A21B)
- Kimi K2.5 (1T)
- Kimi K2.7 Code (1T)
- DeepSeek V4 Flash (284B-A13B)
- Ornith 1.5 397B-A17B
- Qwen3.6 27B
- DeepSeek V3.2 (671B)
- GLM-4.7 (355B-A32B)
- MiniMax M2.7 (230B-A10B)
- Nemotron 3 Ultra 550B-A55B
- DeepSeek R1 0528 (671B)
- Kimi K2 Thinking (1T)
- DeepSeek V3.2 Exp
- Gemma 4 26B-A4B
- MiniMax M2.5 (230B-A10B)
- DeepSeek R1 (671B)
- DeepSeek V4 Flash (April 2026)
- Falcon 180B Chat
- Qwen3.5 122B-A10B
- Step 3.5 Flash (196B-A11B)
- DeepSeek V3.1 Terminus
- DeepSeek V3.1 (671B)
- Gemma 4 31B
- Kimi K2 Instruct 0905 (1T)
- MiniMax M2.1 (230B-A10B)
- Qwen3 235B-A22B Thinking 2507
- Qwen3 Coder 480B-A35B
- MiMo V2 Flash (309B-A15B)
- Kimi K3 (2.8T)
- Ling 1T
- Mistral Large 3 (675B)
- Qwen3 235B-A22B Instruct 2507
- Solar Open 2 250B
- Apodex 1.1 Mini (35B-A3B)
- Kimi K2 Instruct (July 2025)
- MiniMax M2 (230B-A10B)
- Qwen3-VL 235B-A22B Instruct
- Fara 1.5 27B
- GLM-4.7 Flash (30B-A3B)
- K-EXAONE 236B-A23B
- LongCat Flash Chat (560B)
- Nemotron 3 Super 120B-A12B
- Qwen3 235B-A22B
- Agents-A1 35B-A3B
- Qwen AgentWorld 35B-A3B
- Devstral 2 (123B)
- InternVL3.5 241B-A28B
- Laguna S 2.1 (118B-A8B)
- Llama 3.1 Nemotron Ultra 253B v1
- Qwen3 Coder Next (80B-A3B)
- Qwen3-Next 80B-A3B Thinking
- UI-Mate 27B
A
Great · 70+ of 100
Show or hide the list
- Baichuan M3 235B-A22B
- GLM-4.6 (355B-A32B)
- gpt-oss 120B
- Qwen3 Next 80B-A3B Instruct
- DeepSeek V3 0324 (671B)
- Gemma 4 12B
- GLM-4.6V
- MiniMax-M1 80K
- Command A Reasoning (111B)
- DeepSeek R1 Distill Llama 70B
- EXAONE 4.5 33B
- GLM-4.5 (355B-A32B)
- Leanstral 119B-A6B
- Ling 3.0 Flash (127B-A5B)
- Llama 2 70B Chat
- Llama 3.3 70B Instruct
- MiniMax-M1 40K
- Mistral Small 4 (119B-A6B)
- Mistral Medium 3.5 (128B)
- Trinity Large Thinking (400B-A13B)
- Command A (111B)
- ERNIE 4.5 300B-A47B
- GLM-4.5V
- Hermes 4 405B
- Kimi Dev 72B
- Llama 3.3 Nemotron Super 49B v1.5
- Mistral Large 2.1 (2411)
- North Mini Code 1.0 (30B-A3B)
- Seed-OSS 36B Instruct
- Solar Open 100B
- Hermes 4.3 36B
- DiffusionGemma 26B-A4B IT
- GLM-4.5 Air (106B-A12B)
- Mistral Large 2 (123B)
- Pixtral Large 124B (2411)
- Qwen3 30B-A3B Thinking 2507
- Qwen3 Coder 30B-A3B
- Qwen3-VL 32B Instruct
- QwQ 32B
- Ring Flash 2.0 (100B-A6B)
- Nemotron 3 Nano 30B-A3B
- Qwen3 30B-A3B Instruct 2507
- Devstral Small 2 (24B)
- dots.llm1 Instruct
- Granite 4.2 30B
- InternVL3.5 38B
- Ling Flash 2.0 (100B-A6B)
- MiniMax-Text-01
- Olmo 3.1 32B Think
- OpenReasoning Nemotron 32B
- Qwen3 32B
- Qwen3-VL 30B-A3B Thinking
- Qwen3.5 9B
- Skywork OR1 32B
- Tülu 3 405B
- DeepSeek V3 (671B)
- Sarvam 105B
- DeepSeek Coder V2 (236B)
- EXAONE Deep 32B
- GLM-4 32B 0414
- GLM-Z1 32B (0414)
- Hermes 4 70B
- Hunyuan A13B (80B-A13B)
- InternVL3 78B
- Kimi Linear 48B-A3B
- Qwen3 30B-A3B
- Qwen3-VL 30B-A3B Instruct
- Qwen2.5 72B Instruct
- Qwen2.5-VL 72B Instruct
- Apriel 1.5 15B Thinker
- DeepSeek Coder V2 Instruct (0614)
- DeepSeek R1 Distill Qwen 32B
- DeepSeek V2.5 1210
- EXAONE 4.0 32B
- Gemma 3 27B IT
- Granite 4.1 30B
- Llama 3.1 405B Instruct
- Llama 4 Maverick (17B x 128E)
- Llama 4 Scout 17B-16E
- Magistral Small 1.2 (24B)
- Nemotron 3.5 Lightning 30B-A3B
- Olmo 3.1 32B Instruct
- Phi-4 Reasoning Plus (14B)
- Qwen3.6 35B-A3B
- Tülu 3 70B
- Nemotron Cascade 2 30B-A3B
- Ornith 1.0 35B-A3B
- Ornith 1.5 35B-A3B
- Qwen3.5 35B-A3B
- Baichuan M2 32B
- Command R+ (08-2024) 104B
- DeepSeek V2.5
- Devstral Small 1.1 (24B)
- ERNIE 4.5 21B-A3B Thinking
- gpt-oss 20B
- InternVL3 38B
- Jamba 1.5 Large
- Laguna XS 2.1 (33B-A3B)
- Llama 3.1 70B Instruct
- Magistral Small 2507
- MedGemma 27B Text IT
- OpenHands LM 32B
- Phi-4 Reasoning (14B)
- Phi-4 Reasoning Vision 15B
- Qwen2.5 32B Instruct
- Qwen3-Omni 30B-A3B Instruct
- Qwen3.5 27B
- Qwen3.8 27B
- Sarvam 30B
- TMax 27B
- UI-Mate 9B
- UI-Venus 2 9B
B
Good · 55+ of 100
Show the list
- Gemma 4 E4B
- Granite 4.2 8B
- Hermes 3 (Llama 3.1 70B)
- Magistral Small 2506
- Mistral Small 3.2 24B
- QVQ 72B Preview
- Qwen2.5-VL 32B Instruct
- Qwen3 14B
- Qwen2.5 Coder 32B Instruct
- AceReason Nemotron 14B
- Apertus v1.5 70B
- Apriel Nemotron 15B Thinker
- Devstral Small 2505
- Hermes 4 14B
- Llama 3.1 Nemotron 70B Instruct
- Llama 3.2 90B Vision Instruct
- Ministral 3 14B Reasoning
- Mistral Small 3.1 24B
- OpenReasoning Nemotron 14B
- Qwen2.5 Coder 14B Instruct
- QwQ 32B Preview
- Reka Flash 3.1 (21B)
- Trinity Mini (26B-A3B)
- Command R+ 104B (04-2024)
- DeepCoder 14B Preview
- ERNIE 4.5 21B-A3B
- Falcon-H1 34B Instruct
- InternVL3 14B
- Ministral 3 14B
- Mistral Small 3 24B (2501)
- Phi-4 (14B)
- Qwen2 72B Instruct
- Qwen2-VL 72B Instruct
- Codestral 22B
- Command R (08-2024) 32B
- DeepSeek R1 0528 Qwen3 8B
- DeepSeek R1 Distill Qwen 14B
- DeepSeek V2 Chat
- Dolphin Mistral 24B Venice Edition
- Gemma 2 27B IT
- Gemma 4 E2B
- Granite 4.0 H Small (32B-A9B)
- InternVL3.5 8B
- JanusCoder 14B
- Qwen2.5 14B Instruct
- Qwen3-VL 8B Instruct
- Qwen3.5 4B
- Voxtral Small 24B
- TMax 4B
- Agents-A1 4B
- Apertus 70B Instruct
- Aya Expanse 32B
- Aya Vision 32B
- Dolphin 3.0 R1 Mistral 24B
- Gemma 3 12B IT
- GLM-4.1V 9B Thinking
- Granite 4.1 8B
- Kimi-VL A3B Thinking 2506
- Ling 3.0 Tiny (8B-A1.3B)
- MiniCPM-V 4.5 8B
- Nemotron Nano 9B v2
- OLMo 2 32B Instruct
- Qwen2.5 14B Instruct 1M
- Qwen3 8B
- Solar Pro Preview 22B
- Fara 1.5 4B
- GLM-Z1 9B (0414)
- Llama 3 70B Instruct
- MiMo-VL 7B RL
- Qwen2.5-Math 72B Instruct
- Seed-Coder 8B Reasoning
- EXAONE Deep 7.8B
- Falcon-H1R 7B
- InternVL3 8B
- LFM2 24B-A2B
- Mage-VL 4.7B
- Ministral 3 8B
- Phi-3.5 MoE (42B-A6.6B)
- Qwen3 4B Thinking 2507
- Seed-Coder 8B Instruct
- Mixtral 8x22B Instruct
- Qwen2.5 Coder 7B Instruct
C
Fair · 40+ of 100
Show the list
- Command R 35B (v01)
- Llama 3.1 Nemotron Nano 8B v1
- MiniCPM4 8B
- Mistral Small 22B (2409)
- Olmo 3 7B Think
- OpenReasoning Nemotron 7B
- Qwen3 4B Instruct 2507
- Qwen3-VL 4B Instruct
- DeepSeek Coder V2 Lite (16B)
- Falcon 3 10B Instruct
- GLM-4 9B 0414
- Granite 4.2 3B
- Hunyuan 7B Instruct
- LFM2.5 8B-A1B
- Ling Mini 2.0 (16B-A1.4B)
- MiniCPM-o 2.6 8B
- Mistral Nemo 12B Instruct
- Nous Hermes 2 Mixtral 8x7B DPO
- Qwen1.5 110B Chat
- Qwen2.5 7B Instruct
- Qwen2.5-VL 7B Instruct
- Aya 23 35B
- DeepSeek R1 Distill Llama 8B
- Falcon-H1 7B Instruct
- Fara 7B
- Gemma 2 9B IT
- Grok-1 (314B)
- InternLM2.5 20B Chat
- InternLM3 8B Instruct
- InternVL2.5 8B
- Jamba 1.5 Mini
- Kimi-VL A3B Instruct
- Mixtral 8x7B Instruct
- Nemotron 3 Nano 4B
- Olmo 3 7B Instruct
- Pixtral 12B (2409)
- Qwen2 57B-A14B Instruct
- Qwen2.5 7B Instruct 1M
- Qwen3 4B
- Snowflake Arctic Instruct
- Yi 1.5 34B Chat
- GLM-4 9B Chat (2024)
- GLM-4 9B Chat 1M
- Hermes 3 (Llama 3.1 8B)
- Llama 3.1 8B Instruct
- MiniCPM-V 2.6 8B
- Apertus v1.5 8B
- CodeGeeX4 All 9B
- DeepHermes 3 Llama 3 8B Preview
- DeepSeek R1 Distill Qwen 7B
- Dolphin 3.0 Llama 3.1 8B
- EXAONE 3.5 7.8B
- Falcon 3 7B Instruct
- Gemma 3n E4B
- Granite 3.2 8B Instruct
- Granite 3.3 8B
- Granite 4.1 3B
- Ministral 8B (2410)
- Moonlight 16B-A3B Instruct
- Phi-3 Medium (14B)
- Phi-3 Medium 128K Instruct
- Phi-4 Mini Reasoning (3.8B)
- Qwen2.5-Omni 7B
- Tülu 3 8B
- Gemma 3 4B IT
- Ornith 1.0 9B
- Ornith 1.5 9B
- Aya Vision 8B
- Code Llama 70B Instruct
- Command R7B
- DeepSeek Coder 33B Instruct
- Dolphin 2.9.4 Llama 3.1 8B
- Granite 3.1 8B Instruct
- Granite 4.0 H Tiny (7B-A1B)
- Hermes 2 Pro Llama 3 8B
- InternVL2 8B
- Jamba Reasoning 3B
- LFM2.5 2.6B
- LFM2.5-VL 3B
- Llama 3.1 Nemotron Nano 4B v1.1
- LLaVA 1.6 34B
- MedGemma 4B IT
- Ministral 3 3B
- North Micro Vision
- OLMo 2 13B Instruct
- OpenChat 3.6 8B
- Phi-3 Small 8K Instruct
- Phi-4 Mini (3.8B)
- Phi-4 Multimodal Instruct
- Qwen1.5 32B Chat
- Qwen2 7B Instruct
- Qwen2-VL 7B Instruct
- TMax 9B
- MiMo 7B RL
- Command R7B Arabic
- Hunyuan 4B Instruct
- InternLM2 Chat 20B
- Jamba2 3B
- Llama 3 8B Instruct
- Llama 3.2 11B Vision Instruct
- Mistral-NeMo-Minitron 8B Instruct
- SmolLM3 3B
- AFM 4.5B
- Apertus 8B Instruct
- Aya Expanse 8B
- DeepSeek LLM 67B Chat
- DeepSeek Prover V2 7B
- DeepSeek V2 Lite Chat
- Granite 3.0 8B Instruct
- Granite 4.0 Micro (3B)
- Granite Code 34B Instruct
- InternLM2 Math Plus 20B
- InternLM2.5 7B Chat
- Nous Hermes 2 SOLAR 10.7B
- Phi-3.5 Mini Instruct
- Qwen3.5 2B
- SOLAR 10.7B Instruct
- StarCoder2 15B Instruct
- TMax 2B
- Unlimited OCR (3B-A0.6B)
- Yi 34B Chat
- Yi Coder 9B Chat
D
Basic · 0+ of 100
Show the list
- Code Llama 34B Instruct
- CogVLM2 Llama 3 Chat 19B
- Dolphin 2.9 Llama 3 8B
- EXAONE 3.0 7.8B Instruct
- Falcon3 Mamba 7B Instruct
- Granite 4.0 H Micro
- LFM2 8B-A1B
- MiniCPM3 4B
- Mistral 7B Instruct v0.3
- Phi-3.5 Vision Instruct
- Zamba2 7B Instruct
- Codestral Mamba 7B
- Falcon-H1 3B Instruct
- Gemma 3n E2B
- Granite Code 20B Instruct
- Hermes 2 Pro Mistral 7B
- OLMo 2 7B Instruct
- Phi-3 Mini 128K Instruct
- Phi-3 Mini 4K Instruct
- Qwen2.5 Coder 3B Instruct
- Qwen2.5-Math 7B Instruct
- Qwen3-VL 2B Instruct
- Starling LM 7B Beta
- Voxtral Mini 3B
- Yi 1.5 9B Chat
- Llama 3.2 3B Instruct
- OpenChat 3.5 (0106)
- Aya 23 8B
- CodeGemma 7B
- DeepSeek Coder 6.7B Instruct
- LFM2 2.6B
- Mathstral 7B
- Mistral 7B Instruct v0.2
- Nous Hermes 2 Mistral 7B DPO
- OpenChat 3.5 (1210)
- Phi-3 Vision 128K Instruct
- Qwen1.5 14B Chat
- Qwen2.5 3B Instruct
- Qwen2.5-VL 3B Instruct
- RecurrentGemma 9B IT
- Granite Code 8B Instruct (128K)
- Hermes 3 Llama 3.2 3B
- InternLM2 Math Plus 7B
- LFM2.5 1.2B Instruct
- OpenHermes 2.5 Mistral 7B
- StarCoder2 15B (base)
- Starling LM 7B Alpha
- Code Llama 13B Instruct
- EXAONE 3.5 2.4B
- EXAONE 4.0 1.2B
- Falcon 3 3B Instruct
- Falcon Mamba 7B Instruct
- Granite 3.2 2B Instruct
- Granite 3.3 2B
- Mistral 7B Instruct v0.1
- Qwen2.5-Omni 3B
- Qwen3 1.7B
- SQLCoder 7B v2
- Gemma 1.1 7B IT
- Gemma 2 2B
- Granite 3.1 2B Instruct
- Granite 4.0 H 1B
- LLaVA 1.6 Mistral 7B
- Neural Chat 7B v3.3
- Orca 2 13B
- StableLM 2 12B Chat
- Yi 1.5 6B Chat
- Baichuan 2 13B Chat
- DeepScaleR 1.5B Preview
- DeepSeek MoE 16B Chat
- Falcon-H1 1.5B Instruct
- Gemma 7B IT
- Granite 3.0 2B Instruct
- Granite Vision 3.2 2B
- LFM2 1.2B
- Llama 2 13B Chat
- OLMoE 1B-7B Instruct
- Qwen1.5 7B Chat
- Qwen2.5 Coder 1.5B Instruct
- Qwen3.5 0.8B
- StarCoder 15B
- StarCoder2 3B
- WizardMath 7B v1.1
- Zephyr 7B Beta
- Code Llama 7B Instruct
- LFM2-VL 1.6B
- Qwen1.5 MoE A2.7B Chat
- Qwen2-VL 2B Instruct
- Qwen2.5 1.5B Instruct
- SmolVLM2 2.2B Instruct
- StarCoder2 7B
- DeepSeek R1 Distill Qwen 1.5B
- Hunyuan 1.8B Instruct
- Llama 3.2 1B Instruct
- LLaVA 1.5 13B
- Nemotron Mini 4B Instruct
- Orca 2 7B
- SmolVLM 2.2B Instruct
- WizardLM 13B v1.2
- Zamba2 2.7B Instruct
- Baichuan 2 7B Chat
- Gemma 3 1B
- Granite 3.1 3B-A800M Instruct
- Granite Code 3B Instruct (128K)
- Llama 2 7B Chat
- Qwen2.5-Math 1.5B Instruct
- SmolLM2 1.7B Instruct
- Stable Code Instruct 3B
- Vicuna 13B v1.5
- Yi 6B Chat
- DeepSeek LLM 7B Chat
- Granite 3.0 3B-A800M Instruct
- InternLM2.5 1.8B Chat
- LFM2 700M
- Falcon 3 1B Instruct
- Falcon 40B Instruct
- LLaVA 1.5 7B
- Phi-2 2.7B
- Qwen2 1.5B Instruct
- StableLM Zephyr 3B
- Yi Coder 1.5B Chat
- CodeGemma 2B
- Falcon-H1 0.5B Instruct
- LFM2.5 350M
- OLMo 2 1B Instruct
- OLMo 7B Instruct
- Qwen3 0.6B
- SmolLM 1.7B Instruct
- Vicuna 7B v1.5
- DeepSeek Coder 1.3B Instruct
- Gemma 2B IT
- Granite 4.0 H 350M
- InternLM 7B
- LFM2 350M
- LFM2-VL 450M
- LFM2.5 230M
- Qwen2.5 0.5B Instruct
- SmolVLM 500M Instruct
- StableLM 2 1.6B Chat
- StableLM 2 Zephyr 1.6B
- Hunyuan 0.5B Instruct
- Qwen2.5 Coder 0.5B Instruct
- SmolLM2 360M Instruct
- ERNIE 4.5 0.3B
- Falcon 7B Instruct
- Gemma 3 270M
- Qwen2 0.5B Instruct
- SmolVLM 256M Instruct
- TinyLlama 1.1B Chat
- SmolLM 360M Instruct
- SmolLM2 135M Instruct
- Falcon-H1 Tiny 90M Instruct
- SmolLM 135M Instruct
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.