Nukesor/pueue | ProgramBench

:stars: Manage your shell commands.

638

Generated Behavioral Tests

98.6%

Best Score

Claude Opus 5.5 (xhigh)

Results by Model

Hover for details · Click a model to open it · Drag to zoom the x-axis · Shift-drag to pan · The line marks the Pareto frontier (best score per cost)

30 runs

#	Model	Score	Cost	Calls
1	Claude Opus 5.5 (xhigh)	98.6%	$26.53	307	trace →
2	Claude Fable 5.1 (xhigh)	96.2%	$64.65	389	trace →
3	Claude Opus 5 (xhigh)	93.1%	$54.76	334	trace →
4	Muse Wonder Full	85.3%	TBA	658	trace →
5	Muse Wonder (xhigh)	50.8%	$47.24	506	trace →
6	GPT 6 Astra (max)	22.9%	$13.25	63	trace →
7	Claude Opus 4.8 (xhigh)	21.3%	$17.68	119	trace →
8	Muse Spark 1.3 (max)	19.1%	$15.50	496	trace →
9	GLM-5.2	18.0%	$30.03	310	trace →
10	Muse Spark 1.1 (xhigh)	16.8%	$0.45	150	trace →
11	Claude Opus 4.7 (xhigh)	16.1%	$7.48	127	trace →
12	Muse Spark 1.2 (xhigh)	15.8%	$0.95	129	trace →
13	Claude Opus 4.6	15.4%	$12.90	252	trace →
14	GPT 5.5	13.6%	$1.09	14	trace →
15	Gemini 3.7 Flash	13.5%	$2.04	150	trace →
16	Gemini 3.1 Pro	12.5%	$1.07	72	trace →
17	Claude Opus 4.7	12.4%	$1.93	68	trace →
18	Gemini 3.5 Flash	10.8%	$3.76	134	trace →
19	Muse Spark 1.3 (xhigh)	8.8%	$4.77	376	trace →
20	GPT 5.5 (high)	8.6%	$3.36	45	trace →
21	GPT 5.5 (xhigh)	8.3%	$8.76	87	trace →
22	GPT-5.6 Sol (xhigh)	7.8%	$3.36	28	trace →
23	GPT-5.6 Sol	6.4%	$0.71	16	trace →
24	Claude Haiku 4.5	4.2%	$0.56	118	trace →
25	GPT 5.4 mini	4.1%	$0.04	9	trace →
26	GPT 5.4	3.9%	$0.16	7	trace →
27	Gemini 3 Flash	2.8%	$0.30	91	trace →
28	GPT 5 mini	1.9%	$0.02	11	trace →
29	Gemini 3.6 Flash	1.6%	$2.81	105	trace →
30	Claude Sonnet 4.6	1.6%	$15.46	231	trace →

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