Run large language models locally on Mac, Linux, or Windows with a single command.
What They Do
Ollama bundles model weights, quantisation, and a local REST server into a single desktop application. Users run `ollama pull llama3` and immediately get a local API on localhost:11434 compatible with the OpenAI chat format.
Mission
Make running powerful AI models locally as easy as running any other software.
Available Models
| Model | Family | Context | Input /M | Output /M |
|---|---|---|---|---|
| beam | beam | — | — | — |
| clef | clef | — | — | — |
| cogito-2.1 | cogito-2.1 | — | — | — |
| deepseek-v3.1 | deepseek-v3.1 | — | — | — |
| deepseek-v3.2 | deepseek-v3.2 | — | — | — |
| deepseek-v4-flash | deepseek-v4-flash | — | — | — |
| deepseek-v4-pro | deepseek-v4-pro | — | — | — |
| deepseek-v4.1-flash | deepseek-v4.1-flash | — | — | — |
| devstral-2 | devstral-2 | — | — | — |
| devstral-small-2 | devstral-small-2 | — | — | — |
| gemini-3-flash-preview | gemini-3-flash-preview | — | — | — |
| gemma3 | gemma3 | — | — | — |
| gemma4 | gemma4 | — | — | — |
| glm-4.6 | glm-4.6 | — | — | — |
| glm-4.7 | glm-4.7 | — | — | — |
| glm-5 | glm-5 | — | — | — |
| glm-5.1 | glm-5.1 | — | — | — |
| glm-5.2 | glm-5.2 | — | — | — |
| glm-5.3 | glm-5.3 | — | — | — |
| glm-5.3-flash | glm-5.3-flash | — | — | — |
| gpt-oss | gpt-oss | — | — | — |
| kimi-k2 | kimi-k2 | — | — | — |
| kimi-k2-thinking | kimi-k2-thinking | — | — | — |
| kimi-k2.5 | kimi-k2.5 | — | — | — |
| kimi-k2.6 | kimi-k2.6 | — | — | — |
| kimi-k2.7-code | kimi-k2.7-code | — | — | — |
| kimi-k3 | kimi-k3 | — | — | — |
| minimax-m2 | minimax-m2 | — | — | — |
| minimax-m2.1 | minimax-m2.1 | — | — | — |
| minimax-m2.5 | minimax-m2.5 | — | — | — |
| minimax-m2.7 | minimax-m2.7 | — | — | — |
| minimax-m3 | minimax-m3 | — | — | — |
| minimax-m3.1-flash | minimax-m3.1-flash | — | — | — |
| ministral-3 | ministral-3 | — | — | — |
| mistral-large-3 | mistral-large-3 | — | — | — |
| mistral-large-4 | mistral-large-4 | — | — | — |
| muse-spark-open | muse-spark-open | — | — | — |
| nemotron-3-nano | nemotron-3-nano | — | — | — |
| nemotron-3-super | nemotron-3-super | — | — | — |
| nemotron-3-ultra | nemotron-3-ultra | — | — | — |
| qwen3-coder | qwen3-coder | — | — | — |
| qwen3-coder-next | qwen3-coder-next | — | — | — |
| qwen3-next | qwen3-next | — | — | — |
| qwen3-vl | qwen3-vl | — | — | — |
| qwen3.5 | qwen3.5 | — | — | — |
| rnj-1 | rnj-1 | — | — | — |
FAQ
Yes. Ollama itself is open source under MIT licence. The model weights it downloads are subject to each model's individual licence.
Yes. Since Ollama v0.1.24 the /api/chat endpoint accepts the same JSON format as OpenAI's chat.completions, making it a drop-in local replacement for most apps.