It's an interesting experiment but note that it'soptimized for these benchmarks.

They describe it quite clear themselves in a warning:

🚨 1.This model was not trained on tool-calling or agent-based programming data. We therefore do not recommend using it for tasks that involve function calling, API orchestration, or autonomous coding agents. For programming tasks, we recommend using this model on competitive programming problems (e.g., LeetCode-style).

Thinkin’ bout those vibes