wfxr/csview | ProgramBench

📠 Pretty and fast csv viewer for cli with cjk/emoji support.

335

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%	$12.74	224	trace →
2	Muse Wonder Full	100.0%	TBA	265	trace →
3	Claude Fable 5.1 (xhigh)	99.7%	$19.04	166	trace →
4	Muse Wonder (xhigh)	99.7%	$19.04	241	trace →
5	GPT 6 Astra (max)	99.7%	$7.81	54	trace →
6	Claude Opus 5 (xhigh)	99.7%	$25.22	208	trace →
7	Muse Spark 1.3 (max)	99.7%	$3.95	211	trace →
8	Muse Spark 1.3 (xhigh)	99.7%	$0.96	184	trace →
9	GPT 5.5 (xhigh)	99.7%	$4.82	62	trace →
10	GPT-5.6 Sol (xhigh)	99.4%	$2.19	14	trace →
11	Claude Opus 4.8 (xhigh)	99.4%	$14.02	129	trace →
12	GLM-5.2	98.8%	$13.13	158	trace →
13	Muse Spark 1.2 (xhigh)	98.8%	$0.88	188	trace →
14	Gemini 3.7 Flash	98.5%	$1.11	137	trace →
15	Muse Spark 1.1 (xhigh)	98.5%	$0.14	103	trace →
16	Claude Opus 4.7 (xhigh)	98.2%	$11.47	167	trace →
17	GPT 5.5 (high)	97.3%	$2.55	35	trace →
18	GPT-5.6 Sol	97.3%	$0.87	12	trace →
19	Claude Opus 4.6	96.1%	$8.20	249	trace →
20	Gemini 3.6 Flash	94.6%	$3.23	119	trace →
21	GPT 5.5	93.1%	$0.75	12	trace →
22	Gemini 3.1 Pro	93.1%	$1.37	136	trace →
23	Claude Opus 4.7	92.5%	$3.38	68	trace →
24	Claude Sonnet 4.6	92.5%	$7.72	285	trace →
25	Gemini 3.5 Flash	92.2%	$2.72	109	trace →
26	GPT 5.4	88.7%	$0.16	8	trace →
27	Gemini 3 Flash	83.0%	$0.28	100	trace →
28	GPT 5.4 mini	48.1%	$0.05	9	trace →
29	Claude Haiku 4.5	0.6%	$0.63	109	trace →
30	GPT 5 mini	0.0%	$0.02	14	trace →

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