Ollama

Download and run Llama, Mistral, and other models locally with a simple CLI and API.

The default local inference stack — zero config, huge model catalog, perfect dev companion.

الميزات الرئيسية

• Run Llama, Mistral, Gemma locally with one command
• OpenAI-compatible REST API on localhost:11434
• Model library with quantized variants for consumer GPUs
• Cross-platform: macOS, Linux, Windows (WSL)

حالات الاستخدام

1. 01Local LLM prototyping before sending traffic to cloud APIs
2. 02Privacy-sensitive workflows (legal, health, internal docs)
3. 03Pair with Open WebUI or LiteLLM for production-style routing

بدء سريع

curl -fsSL https://ollama.com/install.sh | sh
ollama run llama3.2

توافق المكدس

الوسوم

• #llm
• #self-hosted
• #open-source