#		Model	Score help_outline Percentage of hidden behavioral tests passed.	Cost help_outline Total API cost in USD for this task instance.	Calls help_outline Number of LLM API calls for this task instance.
1		Claude Opus 4.7 Anthropic	56.1%	$16.11	239
2		Claude Sonnet 4.6 Anthropic	42.2%	$45.93	495
3		Claude Opus 4.6 Anthropic	42.0%	$13.07	192
4		Gemini 3 Flash Google	33.3%	$0.31	68
5		Claude Haiku 4.5 Anthropic	30.0%	$1.24	162
6		Gemini 3.1 Pro Google	26.0%	$2.21	100
7		GPT 5.4 OpenAI	25.2%	$0.43	15
8		GPT 5.4 mini OpenAI	14.1%	$0.05	37
9		GPT 5 mini OpenAI	4.6%	$0.01	12

Click row to see model details