Llama 3
Flagship general‑purpose model. Great balance of quality and speed.
ollama pull llama3
Learn more →
Models Hub
Pick a model, copy the pull command, and start your local AI in seconds. Works with the official Ollama app on Windows.
Mistral
Fast, lightweight models for everyday tasks and coding.
ollama pull mistral
Learn more →
Qwen 2.5
Strong reasoning and multilingual capabilities.
ollama pull qwen2.5
Learn more →
Gemma 2
Compact, capable model from Google — efficient on smaller GPUs.
ollama pull gemma2
Learn more →
Phi‑4
Small, instruction‑tuned model — great for edge and quick replies.
ollama pull phi4
Learn more →
More models
Explore additional community models and variants via the CLI.
ollama list
FAQ →
Tip: Model size and speed vary by parameter count and quantization. For best performance, see GPU Acceleration.
Community‑driven guide. Not affiliated with the official Ollama project.