Benchmarks · math
ProofBench v1.1
Automated theorem proving benchmark
43 models · 44 results · updated 2026-10-29 · vals.ai ↗
Leaderboard
AlephProver
Heavy line: best score at each price. Thin lines join one model's effort levels. Cost uses each model's median subscription plan, or API price with no plan data. 31 more models under All models in settings.
More from this benchmark
1 published measuresTime per task53 results · lower is better
31 more models under All models in settings.
All results
44 rows| # | Model | Effort | ProofBench v1.1 | Subscription $ / task | API $ / task | Time |
|---|---|---|---|---|---|---|
| 1 | AlephProver | 100.0% | $9.35 | $9.35 | 18 min | |
| 2 | Fable 5.1 | max | 100.0% | $0.42 | $2.70 | 8.0 min |
| 3 | Opus 5.5 | max | 100.0% | $0.015 | $0.96 | 5.9 min |
| 4 | Sonnet 5.5 | max | 100.0% | $0.008 | $0.56 | 5.8 min |
| 5 | Opus 5 | max | 99.0% | $0.031 | $1.79 | 9.8 min |
| 6 | Gemini 4 Argon | high | 99.0% | $0.56 | $0.78 | 6.3 min |
| 7 | GPT-6 Astra | max | 99.0% | $0.073 | $1.71 | 5.1 min |
| 8 | GPT-6.1 Sol | max | 99.0% | $0.014 | $0.23 | 4.7 min |
| 9 | Fable 5 | max | 95.0% | $0.59 | $5.23 | 14 min |
| 10 | Haiku 5.5 | max | 89.0% | $0.004 | $0.17 | 7.0 min |
| 11 | Kimi K3 | max | 87.0% | $0.15 | $1.65 | 35 min |
| 12 | Aristotle | 86.0% | ||||
| 13 | GPT-5.6 Sol | max | 83.0% | $0.040 | $1.44 | 8.7 min |
| 14 | GPT-6 Sol | max | 83.0% | $0.015 | $0.48 | 5.1 min |
| 15 | Sonnet 5 | max | 77.0% | $0.024 | $1.37 | 15 min |
| 16 | Hy4 preview | high | 75.0% | $0.25 | $0.34 | 23 min |
| 17 | GPT-5.6 Terra | max | 74.0% | $0.035 | $1.27 | 13 min |
| 18 | MiMo V2.6 Pro | max | 70.0% | $0.21 | $0.25 | 1.0 h |
| 19 | GPT-6 Luna | max | 64.0% | $0.003 | $0.042 | 7.5 min |
| 20 | MiMo V2.6 Flash | high | 63.0% | free | $0.12 | 1.0 h |
| 21 | GPT-5.6 Luna | max | 60.0% | $0.006 | $0.12 | 8.6 min |
| 22 | Gemini 3.7 Flash | high | 58.0% | $0.072 | $0.56 | 7.1 min |
| 23 | Muse Spark 1.3 | max | 58.0% | $0.24 | $0.48 | 9.5 min |
| 24 | Qwen 3.8 Max | max | 58.0% | $0.18 | $1.50 | 31 min |
| 25 | DeepSeek V4 Flash 0731 | high | 56.0% | $0.16 | $0.16 | 16 min |
| 26 | Muse Spark 1.3 | xhigh | 55.0% | $0.22 | $0.43 | 8.9 min |
| 27 | DeepSeek V4.1 Flash | high | 54.0% | $0.009 | $0.13 | 9.1 min |
| 28 | Grok 4.6 | high | 51.0% | $0.085 | $0.82 | 14 min |
| 29 | DeepSeek V4 Pro 0813 | max | 50.0% | $0.26 | $0.26 | 12 min |
| 30 | GLM 5.3 | max | 49.0% | $0.22 | $2.08 | 42 min |