o2sh/onefetch | ProgramBench

Command-line Git information tool

1,166

Generated Behavioral Tests

99.2%

Best Score

Muse Wonder Full

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	Muse Wonder Full	99.2%	TBA	523	trace →
2	Claude Opus 5.5 (xhigh)	98.8%	$45.50	366	trace →
3	GPT 6 Astra (max)	98.8%	$10.12	54	trace →
4	Claude Fable 5.1 (xhigh)	98.6%	$82.65	407	trace →
5	Muse Wonder (xhigh)	98.3%	$32.16	409	trace →
6	Claude Opus 5 (xhigh)	96.7%	$96.04	398	trace →
7	Muse Spark 1.3 (max)	89.4%	$7.85	432	trace →
8	GPT 5.5 (xhigh)	85.1%	$12.76	109	trace →
9	GPT-5.6 Sol (xhigh)	84.3%	$6.98	50	trace →
10	GLM-5.2	83.5%	$62.81	327	trace →
11	Gemini 3.7 Flash	83.0%	$2.67	170	trace →
12	Muse Spark 1.3 (xhigh)	82.8%	$5.87	497	trace →
13	GPT 5.5 (high)	81.9%	$4.06	45	trace →
14	Claude Opus 4.6	81.8%	$16.25	289	trace →
15	Claude Opus 4.8 (xhigh)	79.1%	$21.12	145	trace →
16	Muse Spark 1.2 (xhigh)	78.5%	$1.25	132	trace →
17	GPT-5.6 Sol	73.2%	$0.86	12	trace →
18	GPT 5.5	69.0%	$1.42	16	trace →
19	Claude Opus 4.7 (xhigh)	66.0%	$3.68	108	trace →
20	Claude Opus 4.7	53.6%	$1.84	72	trace →
21	GPT 5.4	52.2%	$0.22	8	trace →
22	Claude Haiku 4.5	35.8%	$0.72	98	trace →
23	GPT 5.4 mini	15.8%	$0.03	7	trace →
24	Gemini 3 Flash	13.6%	$0.61	91	trace →
25	Gemini 3.6 Flash	2.3%	$9.51	212	trace →
26	Gemini 3.5 Flash	2.3%	$8.52	220	trace →
27	Claude Sonnet 4.6	2.3%	$33.45	604	trace →
28	Muse Spark 1.1 (xhigh)	0.0%	$1.87	392	trace →
29	Gemini 3.1 Pro	0.0%	$1.68	89	trace →
30	GPT 5 mini	0.0%	$0.02	13	trace →

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