iframe-proxy | Sunbelt Computer Software

CLI for managing secrets

1,748

Generated Behavioral Tests

93.1%

Best Score

Claude Opus 5 (xhigh)

Results by Model

Hover a point for details · The line marks the Pareto frontier (best score per cost) · Click a point to see model details

21 runs

#	Model	Score	Cost	Calls
1	Claude Opus 5 (xhigh)	93.1%	$35.48	216	trace →
2	Gemini 3.7 Flash	88.2%	$1.62	126	trace →
3	GPT 5.5 (xhigh)	88.0%	$15.51	99	trace →
4	Claude Opus 4.8 (xhigh)	86.4%	$12.86	128	trace →
5	GPT-5.6 Sol (xhigh)	85.5%	$6.80	54	trace →
6	Claude Opus 4.6	82.0%	$11.99	232	trace →
7	GLM-5.2	80.0%	$15.24	155	trace →
8	Claude Opus 4.7 (xhigh)	79.3%	$7.80	140	trace →
9	GPT 5.5 (high)	79.0%	$3.86	46	trace →
10	Gemini 3.6 Flash	78.5%	$3.09	101	trace →
11	GPT-5.6 Sol	72.4%	$0.95	13	trace →
12	GPT 5.5	71.3%	$1.20	18	trace →
13	Gemini 3.5 Flash	65.8%	$5.22	169	trace →
14	Claude Sonnet 4.6	62.1%	$14.17	359	trace →
15	GPT 5.4	56.2%	$0.24	9	trace →
16	Claude Haiku 4.5	43.0%	$0.78	117	trace →
17	Gemini 3 Flash	40.3%	$0.21	77	trace →
18	GPT 5 mini	20.9%	$0.04	24	trace →
19	Gemini 3.1 Pro	19.9%	$1.34	66	trace →
20	GPT 5.4 mini	14.5%	$0.03	9	trace →
21	Claude Opus 4.7	9.4%	$3.95	92	trace →

Click a row to replay how that model rebuilt this program, or the model name to open its full run