mgechev/revive | ProgramBench

🔥 ~6x faster, stricter, configurable, extensible, and beautiful drop-in replacement for golint

727

Generated Behavioral Tests

69.9%

Best Score

Muse Wonder (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	Muse Wonder (xhigh)	69.9%	$40.10	443	trace →
2	Claude Opus 5 (xhigh)	65.3%	$40.71	237	trace →
3	Claude Opus 5.5 (xhigh)	65.1%	$71.76	462	trace →
4	Muse Spark 1.3 (max)	55.6%	$3.43	286	trace →
5	Muse Spark 1.3 (xhigh)	53.1%	$1.00	186	trace →
6	Muse Spark 1.1 (xhigh)	49.8%	$0.83	194	trace →
7	GPT 5.5 (high)	48.3%	$3.70	56	trace →
8	Muse Wonder Full	47.9%	TBA	701	trace →
9	GPT 5.5	47.3%	$1.24	22	trace →
10	Gemini 3.7 Flash	47.2%	$1.71	138	trace →
11	GPT-5.6 Sol (xhigh)	47.0%	$5.78	39	trace →
12	Claude Fable 5.1 (xhigh)	46.8%	$97.86	359	trace →
13	Claude Opus 4.7 (xhigh)	46.4%	$2.67	70	trace →
14	Claude Opus 4.6	46.4%	$10.31	254	trace →
15	GPT 5.5 (xhigh)	45.9%	$6.53	64	trace →
16	GPT 6 Astra (max)	44.7%	$10.16	57	trace →
17	Claude Opus 4.7	43.1%	$2.52	62	trace →
18	Gemini 3.6 Flash	42.0%	$2.59	120	trace →
19	Muse Spark 1.2 (xhigh)	41.1%	$0.32	77	trace →
20	Claude Opus 4.8 (xhigh)	40.0%	$46.19	235	trace →
21	Claude Sonnet 4.6	34.1%	$17.26	476	trace →
22	GPT-5.6 Sol	32.3%	$0.82	16	trace →
23	Gemini 3.5 Flash	29.0%	$5.82	220	trace →
24	Claude Haiku 4.5	25.9%	$0.55	149	trace →
25	Gemini 3 Flash	25.0%	$0.24	61	trace →
26	Gemini 3.1 Pro	24.8%	$1.12	75	trace →
27	GPT 5.4	21.3%	$0.38	15	trace →
28	GLM-5.2	11.6%	$28.36	317	trace →
29	GPT 5.4 mini	6.2%	$0.02	8	trace →
30	GPT 5 mini	4.8%	$0.01	8	trace →

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