Han

Han @tieubao [product]

Local LLM benchmark harness for Ollama models.

#python+bash#tool#active

agent-native: HTTP API, the /mcp server, and the vibe CLI. See how it works.