#		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	95.7%	$5.84	142
2		Claude Opus 4.6 Anthropic	94.6%	$11.60	302
3		Claude Sonnet 4.6 Anthropic	87.6%	$15.93	374
4		GPT 5.4 OpenAI	86.8%	$0.25	10
5		GPT 5 mini OpenAI	80.5%	$0.03	13
6		GPT 5.4 mini OpenAI	69.5%	$0.03	9
7		Claude Haiku 4.5 Anthropic	64.9%	$0.57	107
8		Gemini 3.1 Pro Google	58.7%	$1.87	129
9		Gemini 3 Flash Google	3.3%	$0.33	114

Click row to see model details