Hi, thank you very much for open-sourcing this project and sharing your work with the community. I’m currently studying the evaluation of Kimi-K3 MathVision with the Python tool, and I’m particularly interested in how the Python tool is designed and integrated. Could you please share some details about its interface, execution environment, prompting strategy, and how the model decides when to invoke it and use the returned results? If there is any relevant code, documentation, or configuration that I may have overlooked, I would greatly appreciate a pointer. I sincerely appreciate your time and any guidance you can provide!
Hi, thank you very much for open-sourcing this project and sharing your work with the community. I’m currently studying the evaluation of Kimi-K3 MathVision with the Python tool, and I’m particularly interested in how the Python tool is designed and integrated. Could you please share some details about its interface, execution environment, prompting strategy, and how the model decides when to invoke it and use the returned results? If there is any relevant code, documentation, or configuration that I may have overlooked, I would greatly appreciate a pointer. I sincerely appreciate your time and any guidance you can provide!