Comment by hn-aarvin

I'm curious whether token speed played a role here. OpenAI's models may have had an advantage since they tend to run faster than Anthropic's.

A team of GPT-6.1 Sol Ultrafast agents would probably dominate the games.