[jey]

This seems to be incorporated into current LLM generations already -- when code execution is enabled both GPT-5.x and Claude 4.x automatically seem to execute Python code to help with reasoning steps.

[axiom92]

This was integrated in gpt4 2 years ago:

https://www.reddit.com/r/ChatGPT/comments/14sqcg8/anyone_els...

[fsfod]

I remember seeing that GPT-5 had two python tools defined in its leaked prompt one them would hide the output from user visible chain of thought UI.

[Esophagus4]

Same with CoT prompting.

If you compare the outputs of a CoT input vs a control input, the outputs will have the reasoning step either way for the current generation of models.

[logicprog]

Yeah, this is honestly one of the coolest developments of new models.