Ollama

Ollama lets you run large language models locally on your own hardware with zero cost, or use cloud models for faster responses. Download and run Llama, Mistral, Qwen, DeepSeek, and other open source models with a simple CLI. Local models run unlimited and free, keeping your data completely private. Cloud features add the ability to run multiple models simultaneously and access larger models. Integrates with VS Code, Claude Code, LangChain, LlamaIndex, Dify, n8n, and 40,000+ community integrations. OpenAI-compatible API makes it easy to swap into existing applications. Can run fully offline in air-gapped environments.