io.github.Michael-WhiteCapData

ollama-handoff

Offload cheap work from your AI agent to a local Ollama model, at zero cloud cost.

stdiocommunityinfra

Package Details

ollama-handoff

0.1.1•https://pypi.org/project/ollama-handoff
Transportstdio

Environment Variables

OLLAMA_URL
Default:http://localhost:11434

Base URL of the Ollama server.

OLLAMA_DEFAULT_MODEL
Default:qwen2.5-coder:14b

Default model used for handoffs.

OLLAMA_NUM_CTX
Default:32768

Context window in tokens.

OLLAMA_KEEP_ALIVE
Default:30m

How long to keep the model resident in VRAM.

OLLAMA_TIMEOUT_S
Default:600

Per-request timeout in seconds.