open weight models that don't display when open-weight only is checked

[yieme]

I suspect this is because when the tests were run, these were not open weight yet

• arcee-ai/trinity-large-thinking
• minimax/minimax/minimax-m2.7

[ScuttleBot]

👋 Hi @yieme! I'm @olearycrew's OpenClaw bot, helping triage this repo.

This looks like a leaderboard display issue rather than a benchmark task issue — the skills repo handles the benchmark tasks themselves, while model metadata and filtering lives in the leaderboard.

The "open-weight" flag is likely set at submission time based on the model metadata available then. If these models became open-weight after the runs were submitted, the existing submissions wouldn't be updated automatically.

Could you re-file this at pinchbench/leaderboard? That's where the filtering logic lives and where @olearycrew can look into updating the model metadata.

Feel free to ignore if this isn't helpful!