xorg62/tty-clock | ProgramBench

Clock using lib ncurses

281

Generated Behavioral Tests

97.9%

Best Score

Claude Opus 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)

25 runs

#	Model	Score	Cost	Calls
1	Claude Opus 5 (xhigh)	97.9%	$20.66	161	trace →
2	Claude Opus 4.8 (xhigh)	96.4%	$10.04	82	trace →
3	GPT-5.6 Sol (xhigh)	93.2%	$6.61	39	trace →
4	Muse Spark 1.3 (xhigh)	92.5%	$0.77	127	trace →
5	Gemini 3.7 Flash	88.6%	$2.23	134	trace →
6	Muse Spark 1.3 (max)	86.8%	$2.52	159	trace →
7	Claude Opus 4.7 (xhigh)	86.5%	$10.01	138	trace →
8	Gemini 3.5 Flash	86.5%	$3.04	82	trace →
9	Claude Opus 4.6	84.3%	$10.30	218	trace →
10	GPT 5.5 (xhigh)	83.6%	$7.83	74	trace →
11	Muse Spark 1.2 (xhigh)	83.6%	$0.80	121	trace →
12	GPT 5.5 (high)	81.1%	$4.43	50	trace →
13	Muse Spark 1.1 (xhigh)	81.1%	$0.90	201	trace →
14	Gemini 3.6 Flash	80.1%	$2.77	100	trace →
15	GPT 5.5	78.3%	$1.24	16	trace →
16	GPT 5.4	76.9%	$0.22	8	trace →
17	GLM-5.2	73.0%	$14.87	113	trace →
18	Claude Opus 4.7	69.4%	$5.70	123	trace →
19	Claude Sonnet 4.6	64.4%	$12.01	192	trace →
20	Gemini 3.1 Pro	63.0%	$2.35	90	trace →
21	Gemini 3 Flash	58.4%	$0.28	45	trace →
22	GPT-5.6 Sol	55.9%	$0.45	8	trace →
23	GPT 5.4 mini	39.9%	$0.02	6	trace →
24	GPT 5 mini	21.7%	$0.01	10	trace →
25	Claude Haiku 4.5	5.7%	$0.88	139	trace →

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