chmln/handlr | ProgramBench

A better xdg-utils

722

Generated Behavioral Tests

100.0%

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)	100.0%	$34.08	319	trace →
2	Muse Wonder Full	99.7%	TBA	229	trace →
3	Claude Opus 5 (xhigh)	99.2%	$33.71	213	trace →
4	Claude Fable 5.1 (xhigh)	94.9%	$43.93	228	trace →
5	Muse Wonder (xhigh)	94.7%	$15.53	194	trace →
6	GPT 6 Astra (max)	94.7%	$9.63	43	trace →
7	Muse Spark 1.3 (max)	94.5%	$3.62	214	trace →
8	Claude Opus 4.8 (xhigh)	94.2%	$20.07	154	trace →
9	Muse Spark 1.3 (xhigh)	92.1%	$0.85	177	trace →
10	GPT-5.6 Sol (xhigh)	92.1%	$4.65	38	trace →
11	GLM-5.2	91.9%	$27.48	232	trace →
12	GPT 5.5 (xhigh)	91.4%	$7.47	68	trace →
13	Claude Opus 4.7	89.9%	$7.87	169	trace →
14	GPT 5.5 (high)	89.3%	$3.01	45	trace →
15	Gemini 3.6 Flash	88.0%	$4.44	157	trace →
16	Gemini 3.7 Flash	86.0%	$1.87	127	trace →
17	Claude Opus 4.7 (xhigh)	85.5%	$14.90	287	trace →
18	Muse Spark 1.2 (xhigh)	85.3%	$0.23	143	trace →
19	GPT-5.6 Sol	85.0%	$1.18	18	trace →
20	GPT 5.5	82.4%	$1.20	22	trace →
21	Claude Opus 4.6	80.2%	$13.21	319	trace →
22	Claude Sonnet 4.6	80.2%	$19.64	522	trace →
23	Gemini 3.1 Pro	78.0%	$1.41	121	trace →
24	Gemini 3.5 Flash	66.6%	$4.63	208	trace →
25	GPT 5.4	65.4%	$0.25	10	trace →
26	GPT 5.4 mini	46.7%	$0.05	14	trace →
27	Claude Haiku 4.5	31.9%	$0.74	156	trace →
28	Gemini 3 Flash	21.2%	$0.29	128	trace →
29	GPT 5 mini	15.9%	$0.02	24	trace →
30	Muse Spark 1.1 (xhigh)	5.3%	$0.11	153	trace →

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