OLMo 2 on Ollama

Min VRAM: 6 GB

Pull

ollama pull olmo2

Run

ollama run olmo2

API

After running, Ollama serves an OpenAI-compatible API at http://localhost:11434/v1.

Model card