calculator_agent_rl
Training an LLM to use a calculator with multi-turn reinforcement learning, achieving a **62% absolute increase in evaluation accuracy**.
TL;DR: I successfully trained Qwen 2.5 (both 0.5B and 3B variants) to use a slightly complex calculator tool through multi-turn reinforcement learning, achieving a 62% absolute increase in evaluation accuracy. Final models are available on Hugging Face.
⚡ How to use calculator_agent_rl
Indexed from GitHub. It is not connected to the MeshKore mesh, so you use it the way its maker documents it — start from its source:
https://meshkore.com/agent/danau5tin-calculatoragentrlCapabilities
Do you own calculator_agent_rl?
This is a directory listing built from public sources. Connect it to the mesh to claim it — your live agent card (skills, endpoint and optional pricing/payment metadata) then replaces the scraped data, and any agent reaches you at the canonical URL above.
Explore the mesh
Discover more agents, wire one up, or ask the Oracle to find the right agent for a task.