Canop/broot | ProgramBench

A new way to see and navigate directory trees : https://dystroy.org/broot

538

Generated Behavioral Tests

88.1%

Best Score

Claude Fable 5.1 (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 Fable 5.1 (xhigh)	88.1%	$108.94	366	trace →
2	Muse Wonder Full	87.9%	TBA	540	trace →
3	GPT 6 Astra (max)	86.6%	$16.58	77	trace →
4	Claude Opus 5.5 (xhigh)	81.8%	$41.68	317	trace →
5	Muse Wonder (xhigh)	79.9%	$26.38	276	trace →
6	Claude Opus 5 (xhigh)	74.3%	$44.01	191	trace →
7	Gemini 3.7 Flash	68.6%	$2.14	131	trace →
8	Claude Opus 4.6	67.1%	$14.42	241	trace →
9	GPT-5.6 Sol (xhigh)	66.7%	$4.54	25	trace →
10	GPT 5.5 (xhigh)	66.5%	$11.09	83	trace →
11	Muse Spark 1.3 (max)	66.0%	$7.94	343	trace →
12	Muse Spark 1.2 (xhigh)	66.0%	$4.09	408	trace →
13	Claude Opus 4.8 (xhigh)	65.6%	$13.01	110	trace →
14	Muse Spark 1.3 (xhigh)	65.1%	$1.15	196	trace →
15	GLM-5.2	63.9%	$10.67	142	trace →
16	GPT 5.5	62.5%	$1.89	24	trace →
17	Gemini 3.6 Flash	62.3%	$3.63	112	trace →
18	Claude Opus 4.7	61.0%	$1.57	58	trace →
19	Claude Opus 4.7 (xhigh)	59.1%	$6.43	140	trace →
20	Claude Sonnet 4.6	56.5%	$23.30	406	trace →
21	Gemini 3.5 Flash	54.3%	$2.51	86	trace →
22	Gemini 3.1 Pro	52.8%	$2.80	149	trace →
23	GPT 5.5 (high)	52.2%	$4.81	46	trace →
24	GPT-5.6 Sol	42.2%	$1.09	16	trace →
25	Gemini 3 Flash	37.4%	$0.23	42	trace →
26	Claude Haiku 4.5	33.6%	$0.43	99	trace →
27	GPT 5 mini	31.4%	$0.01	9	trace →
28	GPT 5.4	17.8%	$0.42	7	trace →
29	GPT 5.4 mini	7.2%	$0.08	28	trace →
30	Muse Spark 1.1 (xhigh)	5.4%	$1.46	172	trace →

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