cmatsuoka/figlet | ProgramBench

Claudio's FIGlet tree

872

Generated Behavioral Tests

99.8%

Best Score

Claude Fable 5.1 (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 Fable 5.1 (xhigh)	99.8%	$41.35	195	trace →
2	Claude Opus 5.5 (xhigh)	99.5%	$11.10	151	trace →
3	Muse Wonder Full	98.4%	TBA	433	trace →
4	GPT 6 Astra (max)	97.6%	$7.54	43	trace →
5	GPT-5.6 Sol (xhigh)	87.8%	$4.89	25	trace →
6	Gemini 3.7 Flash	86.9%	$6.22	244	trace →
7	Muse Wonder (xhigh)	81.2%	$37.01	370	trace →
8	Muse Spark 1.3 (max)	79.8%	$12.24	333	trace →
9	Gemini 3.6 Flash	79.1%	$26.03	240	trace →
10	Claude Opus 4.8 (xhigh)	78.2%	$15.15	106	trace →
11	Claude Opus 4.6	77.5%	$18.00	279	trace →
12	Claude Opus 4.7 (xhigh)	76.7%	$82.10	502	trace →
13	Muse Spark 1.1 (xhigh)	73.9%	$0.45	282	trace →
14	GPT 5.5 (xhigh)	72.6%	$15.89	137	trace →
15	Claude Opus 4.7	72.2%	$8.06	134	trace →
16	GPT-5.6 Sol	71.8%	$2.57	36	trace →
17	GPT 5.5 (high)	69.6%	$6.70	74	trace →
18	GLM-5.2	68.1%	$16.41	176	trace →
19	Claude Sonnet 4.6	67.0%	$44.94	645	trace →
20	Muse Spark 1.2 (xhigh)	65.3%	$9.04	486	trace →
21	GPT 5.5	45.5%	$2.94	37	trace →
22	GPT 5.4	42.8%	$0.52	16	trace →
23	Gemini 3.1 Pro	40.5%	$1.73	90	trace →
24	Gemini 3.5 Flash	38.8%	$3.26	141	trace →
25	Claude Haiku 4.5	37.4%	$1.43	196	trace →
26	Gemini 3 Flash	30.3%	$0.28	68	trace →
27	GPT 5.4 mini	9.1%	$0.03	10	trace →
28	GPT 5 mini	2.3%	$0.03	17	trace →
29	Claude Opus 5 (xhigh)	0.0%	$14.67	153	trace →
30	Muse Spark 1.3 (xhigh)	0.0%	$2.99	301	trace →

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