ProofBench
As of Nov 3, 2026, ProofBench (v1.1) is maintained by Vals AI. Metric accuracy (%). Ceiling 100. Saturation unknown. Contamination risk low.
- Maintainer
- Vals AI
- Metric
- accuracy (%)
- Ceiling
- 100
- Saturation
- unknown
- Contamination
- low
- Cadence
- Declared cadence: as Vals updates (checked daily). Catalog snapshots: 1, latest Nov 3, 2026.
- Source
- Vals AI · this benchmark on Vals AI
- Grading
- Automatic (tests or exact checks) — Proofs must compile in Lean 4. Saturated at the 100% ceiling, so it is display only.
Source: Vals AI. A proprietary private test set run by Vals; scores are republished with permission. Rows are dated the day isitsota first saw them, and only the benchmark's standard harness counts.
Display only.This benchmark cannot decide a crown under methodology v1.0.
Categories
Where it counts
- Math — display, weight 0
Top 5
Current top scores
- Claude Sonnet 5.5 · 100 · Nov 3, 2026
- Claude Opus 5.5 · 100 · Nov 3, 2026
- Claude Fable 5.1 · 100 · Nov 3, 2026
- GPT-6.1 Sol · 99 · Nov 3, 2026
- GPT-6 Astra · 99 · Nov 3, 2026
| Model | Score | Date | Kind |
|---|---|---|---|
| Claude Sonnet 5.5 | 100 | 2026-11-03 | snapshot |
| Claude Opus 5.5 | 100 | 2026-11-03 | snapshot |
| Claude Fable 5.1 | 100 | 2026-11-03 | snapshot |
| GPT-6.1 Sol | 99 | 2026-11-03 | snapshot |
| GPT-6 Astra | 99 | 2026-11-03 | snapshot |
| Gemini 4 Argon | 99 | 2026-11-03 | snapshot |
| Claude Opus 5 | 99 | 2026-11-03 | snapshot |
| Claude Fable 5 | 95 | 2026-11-03 | snapshot |
| Muse Wonder Full | 94 | 2026-11-03 | snapshot |
| Claude Haiku 5.5 | 89 | 2026-11-03 | snapshot |
| Kimi K3 | 87 | 2026-11-03 | snapshot |
| Muse Wonder | 84 | 2026-11-03 | snapshot |
| GPT-6 Sol | 83 | 2026-11-03 | snapshot |
| GPT-5.6 Sol | 83 | 2026-11-03 | snapshot |
| Claude Sonnet 5 | 77 | 2026-11-03 | snapshot |
| Hy4 Preview | 75 | 2026-11-03 | snapshot |
| GPT-5.6 Terra | 74 | 2026-11-03 | snapshot |
| Mimo V2.6 Pro | 70 | 2026-11-03 | snapshot |
| GPT-6 Luna | 64 | 2026-11-03 | snapshot |
| Mimo V2.6 Flash | 63 | 2026-11-03 | snapshot |
| GPT-5.6 Luna | 60 | 2026-11-03 | snapshot |
| Gemini 3.7 Flash | 58 | 2026-11-03 | snapshot |
| Muse Spark 1.3 | 58 | 2026-11-03 | snapshot |
| Qwen3 8 Max | 58 | 2026-11-03 | snapshot |
| DeepSeek V4 Flash 0731 | 56 | 2026-11-03 | snapshot |
| Muse Spark 1.3 | 55 | 2026-11-03 | snapshot |
| DeepSeek V4.1 Flash | 54 | 2026-11-03 | snapshot |
| Grok 4.6 | 51 | 2026-11-03 | snapshot |
| DeepSeek V4 Pro 0813 | 50 | 2026-11-03 | snapshot |
| GLM 5.3 | 49 | 2026-11-03 | snapshot |