frontier Math

4 models evaluated

#	Model	Provider	Type	Score
1	GPT-5.5 Pro	OpenAI	Closed	52.4%
2	GPT-5.5	OpenAI	Closed	51.7%
3	GPT-5.4 Pro	OpenAI	Closed	50%
4	Claude Opus 4.7 (Adaptive)	Anthropic	Closed	43.8%