OpenAI·released 2026-09-291 source
GPT-6.1 Sol benchmark scores: 15 benchmarks tracked, leading 4. 0% of its scores are independently reproduced — the rest are self-reported or unverified.
Leads on 4 of 15 benchmarks·best result 100 on OTIS Mock AIME 2024-2025 (unverified)·0 of 15 independently reproduced
Head-to-headGPT-6.1 Sol vs Gemini 3.8 FlashLeads on: ARC-AGI, FrontierMath-Tier-4-v2-Private, FrontierMath-Tiers-1-3-v2-Private, OTIS Mock AIME 2024-2025
| Benchmark | Score | Evidence status | Measured |
|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 100accuracy (%) | unverifiedT2 | 2026-09-29 |
| FrontierMath-Tier-4-v2-Private | 100accuracy (%) | unverifiedT2 | 2026-09-29 |
| ProofBench | 99accuracy (%) | unverifiedT1 | 2026-09-29https://www.vals.ai/benchmarks/proof_bench |
| ARC-AGI | 98.5accuracy (%) | unverifiedT1 | 2026-09-29https://arcprize.org/leaderboard |
| ARC-AGI-2 | 94.17accuracy (%) | unverifiedT2 | 2026-09-29 |
| GPQA diamond | 93.86accuracy (%) | unverifiedT2 | 2026-09-29 |
| FrontierMath-Tiers-1-3-v2-Private | 93.68accuracy (%) | unverifiedT2 | 2026-09-29 |
| Mystery Game Puzzles | 77.97accuracy (%) | unverifiedT2 | 2026-09-29 |
| SimpleQA Verified | 73.9accuracy (%) | unverifiedT2 | 2026-09-29 |
| Furniture Assembly | 71.43accuracy (%) | unverifiedT2 | 2026-09-29 |
| Chess Puzzles | 66.33accuracy (%) | unverifiedT2 | 2026-09-29 |
| APEX-Agents | 60accuracy (%) | unverifiedT2 | 2026-09-29 |
| EBR-bench | 54.29accuracy (%) | unverifiedT2 | 2026-09-29 |
| FrontierCode | 50.23accuracy (%) | unverifiedT1 | 2026-09-29https://cognition.com/frontiercode |
| SpireClimb | 24.67accuracy (%) | unverifiedT2 | 2026-09-29 |
Claims drawn from cited facts, not live model generation.
This model was developed by OpenAI and released in 2026.
2 cited facts
It is scored on 15 tracked benchmarks and currently holds top scores on 4 of them. Its record is unverified: outside evaluation has neither confirmed nor disputed the reported results, so the scores should be read as not yet independently validated.
3 cited facts
On average it trails the SOTA leader by 4.7 points under the disclosed harness, a moderate gap that reads as competitive but not front-of-pack. It holds the top score on 4 benchmarks, indicating genuine category leadership on those specific tests rather than overall front-of-pack standing.
2 cited facts
1 facts cross-checked across data sources: 1 single-source.
Source facts, citations, and refresh stamp for this record.
Sources: epoch_benchmarks
GPT-6.1 Sol is an AI model developed by OpenAI, released 2026-09-29. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.
GPT-6.1 Sol has recorded scores on 15 benchmarks, each shown with its evidence status.
GPT-6.1 Sol has recorded scores on 15 benchmarks — GPQA diamond, OTIS Mock AIME 2024-2025, SimpleQA Verified, Chess Puzzles, ARC-AGI, ARC-AGI-2, and 9 more. The full table above shows each score with its evidence status.
0 of 15 recorded scores (0%) are independently reproduced rather than self-reported by the lab.
GPT-6.1 Sol has 15 tracked claims: 15 unverified.