kyoheiu/felix | ProgramBench

tui file manager with vim-like key mapping

501

Generated Behavioral Tests

99.0%

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)	99.0%	$73.54	576	trace →
2	Claude Fable 5.1 (xhigh)	99.0%	$78.97	416	trace →
3	Muse Wonder Full	97.4%	TBA	329	trace →
4	Claude Opus 5 (xhigh)	97.4%	$85.95	384	trace →
5	Muse Spark 1.3 (max)	97.4%	$4.95	278	trace →
6	GPT 6 Astra (max)	97.0%	$6.81	42	trace →
7	Muse Wonder (xhigh)	96.8%	$30.50	395	trace →
8	Claude Opus 4.8 (xhigh)	90.2%	$11.03	121	trace →
9	Muse Spark 1.3 (xhigh)	89.4%	$1.27	237	trace →
10	GPT 5.5 (high)	88.2%	$3.14	51	trace →
11	GPT 5.5	88.2%	$1.12	22	trace →
12	Gemini 3.5 Flash	80.4%	$14.30	200	trace →
13	Claude Opus 4.7 (xhigh)	76.6%	$1.68	52	trace →
14	GPT-5.6 Sol (xhigh)	70.7%	$8.68	65	trace →
15	Muse Spark 1.2 (xhigh)	67.3%	$0.04	40	trace →
16	Gemini 3.6 Flash	65.1%	$4.03	143	trace →
17	GLM-5.2	59.3%	$30.22	247	trace →
18	GPT 5.5 (xhigh)	56.9%	$6.12	46	trace →
19	Gemini 3.1 Pro	49.5%	$1.12	62	trace →
20	GPT 5.4 mini	46.1%	$0.06	44	trace →
21	Claude Opus 4.7	45.3%	$1.13	52	trace →
22	Muse Spark 1.1 (xhigh)	43.7%	$0.26	92	trace →
23	GPT-5.6 Sol	40.3%	$0.70	12	trace →
24	GPT 5.4	39.1%	$0.42	12	trace →
25	Gemini 3 Flash	33.3%	$0.27	76	trace →
26	GPT 5 mini	22.4%	$0.05	20	trace →
27	Claude Haiku 4.5	20.2%	$1.01	142	trace →
28	Gemini 3.7 Flash	1.6%	$1.39	118	trace →
29	Claude Opus 4.6	0.0%	$2.48	244	trace →
30	Claude Sonnet 4.6	0.0%	$4.16	360	trace →

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