Inference
EN
Vals AI

ProofBench v1.1

Automated theorem proving benchmark

ModelsModels with a result.54
Top resultBest result on this benchmark.100.0%Claude Fable 5.2
Top-3 spreadPoints from first to third.0.0 pts
UpdatedDate the source changed the results.9 Nov 2026

Result and cost

52 of 54 results

Results

54 results
#ModelResultThe score from the source.CostUS dollars to run one task.TimeTime to run one task.
01Claude Fable 5.2100.0%$0.7863 min
01Claude Opus 5.5100.0%$0.9246 min
01Claude Sonnet 5.5100.0%$0.5246 min
01Claude Fable 5.1100.0%$2.708 min
01Alephprover100.0%$9.3518 min
06Gemini 4 Argon99.0% ±1.0$0.7836 min
06GPT-6 Astra99.0% ±1.0$1.715 min
06GPT-6.1 Sol99.0% ±1.0$0.2335 min
06Claude Opus 599.0% ±1.0$1.7910 min
10Claude Fable 595.0% ±2.2$5.2314 min
11Muse Wonder Full94.0% ±2.4$0.85612 min
12Claude Haiku 5.589.0% ±3.1$0.1747 min
13Kimi K387.0% ±3.4$1.6535 min
14Aristotle86.0% ±3.5——
15Muse Wonder84.0% ±3.7$1.3115 min
16GPT-5.6 Sol83.0% ±3.8$1.449 min
16GPT-6 Sol83.0% ±3.8$0.4765 min
18Claude Sonnet 577.0% ±4.2$1.3715 min
19Hy4 preview75.0% ±4.4$0.33623 min
20GPT-5.6 Terra74.0% ±4.4$1.2713 min
21MiMo-V2.6-Pro70.0% ±4.6$0.24562 min
22GPT-6 Luna64.0% ±4.8$0.0428 min
23MiMo-V2.6-Flash63.0% ±4.9$0.12363 min
24Ember-161.0% ±4.9$1.7421 min
25GPT-5.6 Luna60.0% ±4.9$0.1239 min
26Gemini 3.7 Flash58.0% ±5.0$0.5597 min
26Qwen3.8 Max58.0% ±5.0$1.5031 min
26Muse Spark 1.3 Max58.0% ±5.0$0.4799 min
29DeepSeek V4 Flash 073156.0% ±5.0$0.15616 min
30Muse Spark 1.355.0% ±5.0$0.4349 min
31DeepSeek V4.1 Flash 101754.0% ±5.0$0.1269 min
32Grok 4.651.0% ±5.0$0.82414 min
33DeepSeek V4 Pro 081350.0% ±5.0$0.26212 min
34GLM-5.349.0% ±5.0$2.0842 min
35Gemini 3.8 Flash48.0% ±5.0$0.6036 min
36Muse Spark 1.243.0% ±5.0$0.4428 min
37Step 5 Preview42.0% ±5.0$0.71539 min
38Gemini 3.5 Flash31.0% ±4.6$0.577 min
38Grok 4.531.0% ±4.6$0.57411 min
40Grok 4.726.0% ±4.4$0.78813 min
40Gemini 3.1 Pro Preview26.0% ±4.4$0.78415 min
42MiMo-V2.5-Pro22.0% ±4.2$0.0920 min
43GLM-5.3-Flash21.0% ±4.1$0.44871 min
44Beam 1.018.0% ±3.9—4 min
44MiniMax M318.0% ±3.9$0.41912 min
46Qwen3.8-27B16.0% ±3.7$2.3024 min
46DeepSeek V4 Pro 042316.0% ±3.7$0.1427 min
46MiMo-V2.516.0% ±3.7$0.05721 min
49Mistral Large 410.0% ±3.0$0.32921 min
50Mistral Medium 3.59.0% ±2.9$1.308 min
51Inkling-Small6.0% ±2.4$0.0828 min
52Mercury 2.53.0% ±1.7$0.0443 min
53Inkling0.0%$0.1997 min
53Nemotron Lightning 3p5 30B A3B0.0%$0.01615 min