Ollama
Install Ollama, pull models, and run them locally. All guides include copy-paste commands.
Install guides
- Run Ollama in Docker (docker pull and docker run)
- Install Ollama on Ubuntu
- Download Ollama for macOS (Apple Silicon and Intel)
- Download and install Ollama on Windows
- Run Ollama on a VPS
Comparisons
Models (28)
| Model | Ollama tag | Min VRAM | Guide |
|---|---|---|---|
| Codestral | codestral | 8 GB | Run guide |
| Command R | command-r | 24 GB | Run guide |
| DeepSeek Coder V2 | deepseek-coder-v2 | 12 GB | Run guide |
| DeepSeek R1 | deepseek-r1 | 8 GB | Run guide |
| Devstral | devstral | 16 GB | Run guide |
| Gemma 3 | gemma3 | 4 GB | Run guide |
| Gemma 4 | gemma2 | 8 GB | Run guide |
| GPT-OSS | gpt-oss | 16 GB | Run guide |
| Granite 4 | granite4 | 4 GB | Run guide |
| Hermes 3 | hermes3 | 6 GB | Run guide |
| Llama 3.1 | llama3.1 | 8 GB | Run guide |
| Llama 3.2 | llama3.2 | 4 GB | Run guide |
| Llama 3.2 Vision | llama3.2-vision | 10 GB | Run guide |
| Llama 3.3 | llama3.3 | 48 GB | Run guide |
| LLaVA | llava | 8 GB | Run guide |
| Mistral | mistral | 8 GB | Run guide |
| Mistral Nemo | mistral-nemo | 8 GB | Run guide |
| Mixtral 8x7B | mixtral | 26 GB | Run guide |
| OLMo 2 | olmo2 | 6 GB | Run guide |
| Phi-3 | phi3 | 4 GB | Run guide |
| Phi-4 | phi4 | 8 GB | Run guide |
| Qwen 2.5 | qwen2.5 | 8 GB | Run guide |
| Qwen2.5 Coder | qwen2.5-coder | 6 GB | Run guide |
| Qwen3 | qwen3 | 8 GB | Run guide |
| Qwen3 Coder | qwen3-coder | 24 GB | Run guide |
| Qwen3.8 27B | qwen3.8:27b | 24 GB | Run guide |
| QwQ | qwq | 24 GB | Run guide |
| SmolLM2 | smollm2 | 3 GB | Run guide |