lfos/calcurse | ProgramBench

A text-based calendar and scheduling application

666

Generated Behavioral Tests

98.8%

Best Score

Muse Wonder Full

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 Full	98.8%	TBA	510	trace →
2	GPT 6 Astra (max)	98.0%	$13.70	71	trace →
3	Muse Wonder (xhigh)	97.4%	$36.67	367	trace →
4	Claude Fable 5.1 (xhigh)	95.3%	$102.50	461	trace →
5	Claude Opus 5.5 (xhigh)	94.3%	$60.83	448	trace →
6	Claude Opus 5 (xhigh)	94.0%	$53.57	339	trace →
7	Muse Spark 1.3 (xhigh)	85.9%	$2.44	266	trace →
8	Claude Opus 4.8 (xhigh)	78.7%	$25.73	175	trace →
9	GPT 5.5 (xhigh)	75.2%	$10.93	86	trace →
10	GLM-5.2	70.6%	$30.69	218	trace →
11	GPT 5.5 (high)	68.8%	$4.46	35	trace →
12	Muse Spark 1.3 (max)	68.2%	$6.62	306	trace →
13	GPT-5.6 Sol	62.8%	$0.95	15	trace →
14	Gemini 3.6 Flash	56.2%	$4.16	166	trace →
15	Gemini 3.5 Flash	56.2%	$11.16	247	trace →
16	GPT 5.5	55.0%	$1.85	27	trace →
17	Claude Opus 4.6	53.8%	$12.74	248	trace →
18	GPT-5.6 Sol (xhigh)	52.7%	$9.25	67	trace →
19	GPT 5.4	35.3%	$0.54	12	trace →
20	Claude Haiku 4.5	35.0%	$0.83	127	trace →
21	Claude Opus 4.7 (xhigh)	33.9%	$3.30	84	trace →
22	Gemini 3 Flash	33.8%	$0.43	113	trace →
23	Claude Opus 4.7	33.5%	$2.88	80	trace →
24	GPT 5.4 mini	26.0%	$0.05	11	trace →
25	Gemini 3.1 Pro	21.3%	$1.69	84	trace →
26	GPT 5 mini	17.0%	$0.01	8	trace →
27	Gemini 3.7 Flash	7.1%	$6.81	255	trace →
28	Muse Spark 1.2 (xhigh)	7.1%	$3.61	223	trace →
29	Muse Spark 1.1 (xhigh)	7.1%	$1.37	172	trace →
30	Claude Sonnet 4.6	7.1%	$31.57	528	trace →

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