cordx56/rustowl | ProgramBench

Visualize Ownership and Lifetimes in Rust

589

Generated Behavioral Tests

99.8%

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)	99.8%	$25.92	271	trace →
2	Muse Wonder (xhigh)	99.2%	$38.23	395	trace →
3	Muse Spark 1.3 (max)	96.9%	$4.23	273	trace →
4	GPT 6 Astra (max)	96.6%	$8.27	50	trace →
5	Claude Fable 5.1 (xhigh)	96.4%	$52.27	273	trace →
6	Muse Wonder Full	96.1%	TBA	352	trace →
7	Claude Opus 5 (xhigh)	95.4%	$38.72	242	trace →
8	Claude Opus 4.8 (xhigh)	93.0%	$8.41	91	trace →
9	Muse Spark 1.3 (xhigh)	92.0%	$0.77	153	trace →
10	Muse Spark 1.1 (xhigh)	90.7%	$0.22	174	trace →
11	Gemini 3.7 Flash	89.0%	$1.14	134	trace →
12	GPT 5.5 (high)	88.5%	$2.86	38	trace →
13	GLM-5.2	87.4%	$11.08	134	trace →
14	Muse Spark 1.2 (xhigh)	85.7%	$0.34	105	trace →
15	GPT 5.5 (xhigh)	78.3%	$5.00	49	trace →
16	GPT-5.6 Sol (xhigh)	76.9%	$3.22	31	trace →
17	Claude Sonnet 4.6	75.2%	$10.43	237	trace →
18	GPT-5.6 Sol	74.7%	$1.09	18	trace →
19	Claude Opus 4.6	71.8%	$10.74	250	trace →
20	GPT 5.5	66.9%	$0.67	14	trace →
21	Claude Opus 4.7 (xhigh)	65.7%	$4.86	122	trace →
22	Gemini 3.5 Flash	64.0%	$8.41	196	trace →
23	Claude Opus 4.7	62.8%	$2.22	84	trace →
24	Gemini 3.1 Pro	56.4%	$1.96	123	trace →
25	Claude Haiku 4.5	44.1%	$0.65	108	trace →
26	GPT 5.4	42.4%	$0.25	10	trace →
27	Gemini 3 Flash	39.2%	$0.29	65	trace →
28	GPT 5 mini	29.2%	$0.02	19	trace →
29	GPT 5.4 mini	25.3%	$0.03	7	trace →
30	Gemini 3.6 Flash	0.0%	$4.17	165	trace →

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