peco/peco | ProgramBench

Simplistic interactive filtering tool

1,215

Generated Behavioral Tests

99.2%

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)	99.2%	$44.84	403	trace →
2	GPT 6 Astra (max)	98.8%	$9.86	54	trace →
3	Muse Wonder Full	98.0%	TBA	384	trace →
4	Muse Wonder (xhigh)	95.5%	$34.45	378	trace →
5	Muse Spark 1.3 (max)	94.2%	$5.10	340	trace →
6	Claude Fable 5.1 (xhigh)	92.3%	$39.22	333	trace →
7	Muse Spark 1.3 (xhigh)	91.4%	$3.11	360	trace →
8	Claude Opus 5 (xhigh)	89.6%	$54.28	389	trace →
9	GLM-5.2	89.6%	$25.62	210	trace →
10	GPT-5.6 Sol (xhigh)	86.0%	$5.75	40	trace →
11	Claude Opus 4.8 (xhigh)	82.8%	$41.44	230	trace →
12	Gemini 3.7 Flash	81.6%	$1.19	118	trace →
13	Claude Opus 4.6	76.8%	$14.19	391	trace →
14	Gemini 3.6 Flash	76.4%	$5.02	181	trace →
15	Gemini 3.5 Flash	76.1%	$3.84	152	trace →
16	Claude Sonnet 4.6	74.7%	$15.96	351	trace →
17	Claude Opus 4.7 (xhigh)	72.3%	$9.13	157	trace →
18	Gemini 3.1 Pro	71.9%	$1.57	74	trace →
19	Muse Spark 1.2 (xhigh)	71.2%	$0.42	86	trace →
20	GPT 5.5 (high)	69.8%	$4.35	46	trace →
21	GPT 5.5 (xhigh)	68.7%	$8.02	84	trace →
22	Claude Opus 4.7	67.6%	$2.50	84	trace →
23	GPT 5.4	66.6%	$0.29	11	trace →
24	GPT 5.5	65.4%	$1.13	16	trace →
25	GPT-5.6 Sol	61.7%	$0.36	7	trace →
26	GPT 5 mini	56.9%	$0.03	10	trace →
27	Claude Haiku 4.5	51.9%	$1.14	146	trace →
28	GPT 5.4 mini	29.0%	$0.03	8	trace →
29	Gemini 3 Flash	11.4%	$0.20	45	trace →
30	Muse Spark 1.1 (xhigh)	3.5%	$0.33	115	trace →

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