[mikeknoop]

Yes there is a secondary leaderboard called ARC-AGI-Pub (in beta) with no limitations: https://arcprize.org/leaderboard

[p1esk]

I don’t see gpt4 scores there. In fact I’m particularly interested in the performance of a natively multimodal model, like gpt4o or gemini. It does not really make sense to test a model trained on text on those visual/spatial puzzles.