single mention · inference & serving
Python execution tool in chain-of-thought
A Python execution tool lets the model run Python code inside its chain of thought for internal reasoning.1 Code sent to the tool runs in a stateful notebook environment, and the tool returns the execution output to the model.2
How this definition was written
Summarised by glm-5.3-flash from 1 source; every sentence is checked against the quote it cites.
Use this tool to execute Python code in your chain of thought. The code will not be shown to the user. This tool should be used for internal reasoning
— OpenAI Harmony Response Formatit will be executed in a stateful Jupyter notebook environment. python will respond with the output of the execution or time out
— OpenAI Harmony Response Format
Also called Python tool.
Filed under inference & serving › agentic scaffolding