Free, open-source runtime for running open LLMs locally on macOS, Windows, and Linux — CLI, OpenAI-compatible API, GGUF import. Local runs unlimited.
Ollama homepage — run open models locally
Overview
What is Ollama?
The most popular way to run open models locally on your own machine.
Ollama is a free, open-source runtime for running open large language models on your own machine (macOS, Windows, Linux). It bundles a command-line interface, a REST API compatible with OpenAI and Anthropic endpoints, and a model library you pull with a single command — plus GGUF import via Modelfiles.
Local runs are always free and unlimited on any plan; prompts and data never leave your machine. Ollama Cloud offers hosted inference for models too large to run locally, with paid plans (Pro $20/month, Team $500/month) covering cloud usage credits.
Platforms and languages
Ollama Availability
Platforms
Languages
Capabilities
Ollama Key Features
One-command local models
Pull and run open models locally with `ollama pull` and `ollama run`.
GGUF import
Bring your own quantized models with a Modelfile — FROM a .gguf file, then `ollama create`.
Compatible APIs
OpenAI-compatible /v1/chat/completions and Anthropic-compatible /v1/messages endpoints for drop-in integration.
Privacy-first
Local runs never send prompts or data to ollama.com; cloud prompts are never stored or trained on.
Ollama Cloud
Run larger models on Ollama's servers via :cloud model names when local hardware isn't enough.
Best for
Who uses Ollama?
Developers and privacy-minded users
Run open LLMs locally with a CLI and OpenAI-compatible API
Plans and access
Ollama Pricing
Freemium
Local runs always free and unlimited. Pro $20/mo, Team $500/mo, Enterprise custom. Cloud credits extra.
Common questions
Ollama FAQs
Does Ollama send my prompts back to ollama.com?
No for local runs. Cloud-hosted prompts are processed to provide the service but never stored, logged, or trained on.
Where are models stored?
macOS ~/.ollama/models, Linux /usr/share/ollama/.ollama/models, Windows C:\Users\%username%\.ollama\models — changeable via the OLLAMA_MODELS environment variable.
Are local runs limited by plan?
No — running models on your own hardware is always unlimited; paid plans only cover cloud usage credits.
How do I import a GGUF model?
Write a Modelfile with a FROM line pointing at the .gguf file, then `ollama create my-model` and `ollama run my-model`.
Reviews
See what the community thinks and share your experience.
—
Based on 0 ratings
Rating distribution
Community reviews
No reviews
No written reviews yet.
