jrnxf/thokr | ProgramBench

✨ sleek typing tui with visualized results and historical logging

440

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%	$18.60	184	trace →
2	Claude Fable 5.1 (xhigh)	100.0%	$42.55	234	trace →
3	Muse Wonder Full	100.0%	TBA	269	trace →
4	GPT 6 Astra (max)	99.5%	$11.29	57	trace →
5	Claude Opus 5 (xhigh)	99.5%	$41.93	202	trace →
6	Muse Wonder (xhigh)	98.9%	$25.65	288	trace →
7	GLM-5.2	96.6%	$17.71	137	trace →
8	Claude Opus 4.8 (xhigh)	95.2%	$14.61	105	trace →
9	GPT-5.6 Sol (xhigh)	94.5%	$3.88	32	trace →
10	Claude Opus 4.7 (xhigh)	93.6%	$18.61	257	trace →
11	GPT 5.5 (xhigh)	92.5%	$6.84	60	trace →
12	Muse Spark 1.3 (xhigh)	91.8%	$1.51	220	trace →
13	GPT-5.6 Sol	88.2%	$0.93	12	trace →
14	GPT 5.5 (high)	84.8%	$2.42	29	trace →
15	GPT 5.5	83.6%	$1.04	19	trace →
16	Claude Opus 4.6	82.3%	$18.03	328	trace →
17	Muse Spark 1.1 (xhigh)	78.4%	$1.50	185	trace →
18	Muse Spark 1.2 (xhigh)	76.6%	$0.40	130	trace →
19	Gemini 3.6 Flash	66.8%	$4.86	159	trace →
20	Claude Opus 4.7	63.9%	$4.85	126	trace →
21	Muse Spark 1.3 (max)	63.6%	$15.28	695	trace →
22	Gemini 3.7 Flash	50.0%	$1.30	112	trace →
23	Gemini 3.5 Flash	46.4%	$4.44	113	trace →
24	GPT 5.4	43.9%	$0.09	6	trace →
25	Gemini 3.1 Pro	42.7%	$1.31	89	trace →
26	Gemini 3 Flash	40.7%	$0.32	96	trace →
27	GPT 5.4 mini	30.2%	$0.02	8	trace →
28	GPT 5 mini	26.6%	$0.02	12	trace →
29	Claude Haiku 4.5	23.0%	$0.61	100	trace →
30	Claude Sonnet 4.6	0.0%	$44.83	668	trace →

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