Which Models Can You Run?
Every model in the catalog, ranked for your hardware. We auto-detect what you're on; pick manually if it's wrong.
Detecting your hardware…
Sort
Grades
105 of 105 models
| Model | Params | Min VRAM | Status | |
|---|---|---|---|---|
| Aya Expanse 32B Cohere · Aya Expanse | 32B | 22.0 GB Q4_K_M | — | View → |
| Aya Expanse 8B Cohere · Aya Expanse | 8B | 6.5 GB Q4_K_M | — | View → |
| Codestral 22B Mistral AI · Mistral | 22B | 14.7 GB Q4_K_M | — | View → |
| Codestral Mamba 7B Mistral AI · Mistral | 7B | 6.9 GB Q4_K_M | — | View → |
| Cogito 70B Deep Cogito · Cogito | 70B | 43.0 GB Q4_K_M | — | View → |
| Cogito 32B Deep Cogito · Cogito | 32B | 21.5 GB Q4_K_M | — | View → |
| Cogito 8B Deep Cogito · Cogito | 8B | 7.5 GB Q4_K_M | — | View → |
| Command A 111B Cohere · Command R | 111B | 61.0 GB Q4_K_M | — | View → |
| Command R 35B Cohere · Command R | 35B | 22.5 GB Q4_K_M | — | View → |
| Command R+ 104B Cohere · Command R | 104B | 57.0 GB Q4_K_M | — | View → |
| DeepSeek R1 1.5B DeepSeek · DeepSeek R1 | 1.5B | 3.0 GB Q8_0 | — | View → |
| DeepSeek R1 14B DeepSeek · DeepSeek R1 | 14B | 9.9 GB Q4_K_M | — | View → |
| DeepSeek R1 32B DeepSeek · DeepSeek R1 | 32B | 20.7 GB Q4_K_M | — | View → |
| DeepSeek R1 671B DeepSeek · DeepSeek R1 | 671B | 362.0 GB Q4_K_M | — | View → |
| DeepSeek R1 7B DeepSeek · DeepSeek R1 | 7B | 9.0 GB Q8_0 | — | View → |
| DeepSeek R1 8B DeepSeek · DeepSeek R1 | 8B | 7.5 GB Q4_K_M | — | View → |
| DeepSeek V3-0324 DeepSeek · DeepSeek V3 | 671B | 362.0 GB Q4_K_M | — | View → |
| DeepSeek V3.2 DeepSeek · DeepSeek V3 | 671B | 420.0 GB Q4_K_M | — | View → |
| DeepSeek R1 70B DeepSeek · DeepSeek R1 | 70B | 43.5 GB Q4_K_M | — | View → |
| DeepSeek V3 DeepSeek · DeepSeek V3 | 671B | 362.0 GB Q4_K_M | — | View → |
| Devstral 2 123B Mistral AI · Mistral | 123B | 67.0 GB Q4_K_M | — | View → |
| Devstral 24B Mistral AI · Mistral | 24B | 17.0 GB Q4_K_M | — | View → |
| Dolphin Mixtral 8x7B Cognitive Computations · Dolphin | 47B | 26.0 GB Q4_K_M | — | View → |
| Dolphin 3 8B Cognitive Computations · Dolphin | 8B | 6.0 GB Q4_K_M | — | View → |
| Falcon 3 10B TII · Falcon 3 | 10B | 8.5 GB Q4_K_M | — | View → |
| Falcon 3 7B TII · Falcon 3 | 7B | 6.8 GB Q4_K_M | — | View → |
| Gemma 2 27B Google · Gemma 2 | 27B | 17.7 GB Q4_K_M | — | View → |
| Gemma 2 2B Google · Gemma 2 | 2B | 4.0 GB Q8_0 | — | View → |
| Gemma 2 9B Google · Gemma 2 | 9B | 11.0 GB Q8_0 | — | View → |
| Gemma 3 12B Google · Gemma 3 | 12B | 10.5 GB Q4_K_M | — | View → |
| Gemma 3 1B Google · Gemma 3 | 1B | 2.0 GB Q8_0 | — | View → |
| Gemma 3 27B Google · Gemma 3 | 27B | 20.0 GB Q4_K_M | — | View → |
| Gemma 3 4B Google · Gemma 3 | 4B | 5.0 GB Q4_K_M | — | View → |
| Gemma 3n E2B Google · Gemma 3 | 2B | 3.3 GB Q4_K_M | — | View → |
| Gemma 3n E4B Google · Gemma 3 | 4B | 4.5 GB Q4_K_M | — | View → |
| Gemma 4 26B Google · Gemma 4 | 26B | 20.0 GB Q4_K_M | — | View → |
| Gemma 4 E2B Google · Gemma 4 | 2B | 4.0 GB Q4_K_M | — | View → |
| Gemma 4 31B Google · Gemma 4 | 31B | 22.0 GB Q4_K_M | — | View → |
| Gemma 4 E4B Google · Gemma 4 | 4B | 6.0 GB Q4_K_M | — | View → |
| GLM-5.1 Zhipu AI · GLM | 754B | 305.0 GB Q2_K | — | View → |
| GLM-5 Zhipu AI · GLM | 744B | 300.0 GB Q2_K | — | View → |
| Granite 3.3 8B IBM · Granite | 8B | 10.0 GB Q8_0 | — | View → |
| InternLM 2.5 20B Shanghai AI Lab · InternLM 2.5 | 20B | 12.0 GB Q4_K_M | — | View → |
| InternLM 2.5 7B Shanghai AI Lab · InternLM 2.5 | 7B | 5.5 GB Q4_K_M | — | View → |
| Kimi K2.5 Moonshot AI · Kimi | 1040B | 390.0 GB Q2_K | — | View → |
| Llama 3.1 405B Meta · Llama 3 | 405B | 244.5 GB Q4_K_M | — | View → |
| Llama 3.1 70B Meta · Llama 3 | 70B | 43.5 GB Q4_K_M | — | View → |
| Llama 3.1 8B Meta · Llama 3 | 8B | 10.0 GB Q8_0 | — | View → |
| Llama 3.2 1B Meta · Llama 3 | 1B | 3.0 GB Q8_0 | — | View → |
| Llama 3.2 3B Meta · Llama 3 | 3B | 5.0 GB Q8_0 | — | View → |
| Llama 3.2 Vision 11B Meta · Llama 3 | 11B | 8.5 GB Q4_K_M | — | View → |
| Llama 3.2 Vision 90B Meta · Llama 3 | 90B | 50.0 GB Q4_K_M | — | View → |
| Llama 3.3 70B Meta · Llama 3 | 70B | 43.5 GB Q4_K_M | — | View → |
| Llama 4 Maverick Meta · Llama 4 | 400B | 228.0 GB Q4_K_M | — | View → |
| Llama 4 Scout (109B/17B active) Meta · Llama 4 | 109B | 72.0 GB Q4_K_M | — | View → |
| Magistral Small 24B Mistral AI · Mistral | 24B | 17.0 GB Q4_K_M | — | View → |
| Mistral 7B Mistral AI · Mistral | 7B | 9.0 GB Q8_0 | — | View → |
| Mistral Large 2 123B Mistral AI · Mistral | 123B | 67.0 GB Q4_K_M | — | View → |
| Mistral Nemo 12B Mistral AI · Mistral | 12B | 9.5 GB Q4_K_M | — | View → |
| Mistral Small 3.1 24B Mistral AI · Mistral | 24B | 18.0 GB Q4_K_M | — | View → |
| Mixtral 8x22B Mistral AI · Mistral | 141B | 86.0 GB Q4_K_M | — | View → |
| Mixtral 8x7B Mistral AI · Mistral | 47B | 29.7 GB Q4_K_M | — | View → |
| Nemotron 3 Nano 8B NVIDIA · Nemotron | 8B | 7.5 GB Q4_K_M | — | View → |
| Nemotron Ultra 253B NVIDIA · Nemotron | 253B | 155.0 GB Q4_K_M | — | View → |
| Nous Hermes 2 34B Nous Research · Nous Hermes | 34B | 19.0 GB Q4_K_M | — | View → |
| Nous Hermes 2 8B Nous Research · Nous Hermes | 8B | 6.0 GB Q4_K_M | — | View → |
| OpenChat 3.5 7B OpenChat · OpenChat | 7B | 6.9 GB Q4_K_M | — | View → |
| Phi-3 Mini 3.8B Microsoft · Phi | 3.8B | 5.8 GB Q8_0 | — | View → |
| Phi-4 14B Microsoft · Phi | 14B | 9.9 GB Q4_K_M | — | View → |
| Phi-4 Mini 3.8B Microsoft · Phi | 3.8B | 4.5 GB Q4_K_M | — | View → |
| Phi-4 Reasoning 14B Microsoft · Phi | 14B | 11.0 GB Q4_K_M | — | View → |
| Qwen 2.5 14B Alibaba · Qwen 2.5 | 14B | 9.9 GB Q4_K_M | — | View → |
| Qwen 2.5 32B Alibaba · Qwen 2.5 | 32B | 20.7 GB Q4_K_M | — | View → |
| Qwen 2.5 72B Alibaba · Qwen 2.5 | 72B | 44.7 GB Q4_K_M | — | View → |
| Qwen 2.5 Coder 14B Alibaba · Qwen 2.5 | 14B | 12.0 GB Q4_K_M | — | View → |
| Qwen 2.5 7B Alibaba · Qwen 2.5 | 7B | 9.0 GB Q8_0 | — | View → |
| Qwen 2.5 Coder 32B Alibaba · Qwen 2.5 | 32B | 23.0 GB Q4_K_M | — | View → |
| Qwen 2.5 Coder 7B Alibaba · Qwen 2.5 | 7B | 9.0 GB Q8_0 | — | View → |
| Qwen 2.5 VL 72B Alibaba · Qwen 2.5 | 72B | 41.0 GB Q4_K_M | — | View → |
| Qwen 2.5 VL 7B Alibaba · Qwen 2.5 | 7B | 7.0 GB Q4_K_M | — | View → |
| Qwen 3 0.6B Alibaba · Qwen 3 | 0.6B | 2.5 GB Q4_K_M | — | View → |
| Qwen 3 235B-A22B Alibaba · Qwen 3 | 235B | 138.0 GB Q4_K_M | — | View → |
| Qwen 3 14B Alibaba · Qwen 3 | 14B | 12.0 GB Q4_K_M | — | View → |
| Qwen 3 30B-A3B (MoE) Alibaba · Qwen 3 | 30B | 22.0 GB Q4_K_M | — | View → |
| Qwen 3 32B Alibaba · Qwen 3 | 32B | 23.0 GB Q4_K_M | — | View → |
| Qwen 3 4B Alibaba · Qwen 3 | 4B | 4.5 GB Q4_K_M | — | View → |
| Qwen 3 8B Alibaba · Qwen 3 | 8B | 7.5 GB Q4_K_M | — | View → |
| Qwen 3.5 0.8B Alibaba · Qwen 3.5 | 0.8B | 1.5 GB Q4_K_M | — | View → |
| Qwen 3.5 122B Alibaba · Qwen 3.5 | 122B | 85.0 GB Q4_K_M | — | View → |
| Qwen 3.5 27B Alibaba · Qwen 3.5 | 27B | 19.0 GB Q4_K_M | — | View → |
| Qwen 3.5 35B A3B Alibaba · Qwen 3.5 | 35B | 12.0 GB Q4_K_M | — | View → |
| Qwen 3.5 2B Alibaba · Qwen 3.5 | 2B | 3.0 GB Q4_K_M | — | View → |
| Qwen 3.5 4B Alibaba · Qwen 3.5 | 4B | 4.5 GB Q4_K_M | — | View → |
| Qwen 3.5 9B Alibaba · Qwen 3.5 | 9B | 7.5 GB Q4_K_M | — | View → |
| SmolLM2 1.7B Hugging Face · SmolLM | 1.7B | 2.7 GB Q8_0 | — | View → |
| QwQ 32B Alibaba · Qwen 3 | 32B | 21.5 GB Q4_K_M | — | View → |
| StarCoder2 15B BigCode · StarCoder | 15B | 17.0 GB Q8_0 | — | View → |
| StarCoder2 3B BigCode · StarCoder | 3B | 3.5 GB Q4_K_M | — | View → |
| StarCoder2 7B BigCode · StarCoder | 7B | 5.5 GB Q4_K_M | — | View → |
| WizardCoder 33B Microsoft · WizardLM | 33B | 22.0 GB Q4_K_M | — | View → |
| WizardLM 2 7B Microsoft · WizardLM | 7B | 6.9 GB Q4_K_M | — | View → |
| Yi 1.5 34B 01.AI · Yi 1.5 | 34B | 21.0 GB Q4_K_M | — | View → |
| Yi 1.5 6B 01.AI · Yi 1.5 | 6B | 5.0 GB Q4_K_M | — | View → |
| Yi 1.5 9B 01.AI · Yi 1.5 | 9B | 6.5 GB Q4_K_M | — | View → |
| Yi Coder 9B 01.AI · Yi Coder | 9B | 8.0 GB Q4_K_M | — | View → |
S · Runs great
Fast inference with plenty of VRAM headroom. Score 85+.
A · Runs well
Solid speed and headroom. Score 70–84.
B · Decent
Usable speed, moderate headroom. Score 55–69.
C · Tight fit
Limited headroom or sluggish speed. Score 40–54.
D · Barely runs
Slow tokens, very tight on VRAM. Score 20–39.
F · Too heavy
Won't load well or at all. Score 0–19.