danmar/cppcheck | ProgramBench

static analysis of C/C++ code

2,126

Generated Behavioral Tests

44.3%

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)	44.3%	$70.42	261	trace →
2	Muse Wonder (xhigh)	38.2%	$33.99	299	trace →
3	Claude Opus 5.5 (xhigh)	37.2%	$38.24	315	trace →
4	Muse Spark 1.3 (max)	28.9%	$4.44	263	trace →
5	Claude Opus 5 (xhigh)	24.8%	$14.91	151	trace →
6	Claude Opus 4.8 (xhigh)	19.3%	$4.94	67	trace →
7	Gemini 3.6 Flash	18.7%	$3.84	122	trace →
8	Gemini 3.5 Flash	18.3%	$4.17	152	trace →
9	Claude Haiku 4.5	14.5%	$0.74	112	trace →
10	GPT 5 mini	13.8%	$0.04	15	trace →
11	Muse Spark 1.3 (xhigh)	10.6%	$0.57	150	trace →
12	Claude Opus 4.6	10.5%	$6.95	175	trace →
13	Gemini 3 Flash	8.7%	$0.20	74	trace →
14	GPT 6 Astra (max)	6.0%	$14.06	79	trace →
15	Muse Wonder Full	5.1%	TBA	753	trace →
16	Muse Spark 1.2 (xhigh)	4.9%	$0.20	97	trace →
17	GPT 5.5 (xhigh)	4.9%	$5.35	63	trace →
18	GLM-5.2	4.8%	$9.20	131	trace →
19	Gemini 3.1 Pro	4.5%	$1.64	94	trace →
20	Claude Opus 4.7 (xhigh)	3.7%	$1.94	71	trace →
21	Claude Sonnet 4.6	3.0%	$15.02	370	trace →
22	Claude Opus 4.7	2.7%	$0.62	27	trace →
23	GPT 5.4	2.3%	$0.24	9	trace →
24	GPT 5.5	2.2%	$0.73	13	trace →
25	GPT-5.6 Sol	2.2%	$0.71	11	trace →
26	GPT 5.5 (high)	2.0%	$2.02	32	trace →
27	GPT 5.4 mini	1.9%	$0.03	11	trace →
28	GPT-5.6 Sol (xhigh)	1.5%	$3.23	24	trace →
29	Gemini 3.7 Flash	0.2%	$0.99	110	trace →
30	Muse Spark 1.1 (xhigh)	0.2%	$0.09	127	trace →

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