OpenAI·released 2024-12-171 source
o1 benchmark scores: 19 benchmarks tracked. 16% of its scores are independently reproduced — the rest are self-reported or unverified.
Leads on 0 of 19 benchmarks·best result 94.71 on MATH level 5 (reproduced)·3 of 19 independently reproduced·$15/$60 per M tokens
Consensus: LiteLLM · Cross-check: models.dev, OpenRouter · See every model’s pricing →
| Benchmark | Score | Evidence status | Measured |
|---|---|---|---|
| MATH level 5 | 94.71accuracy (%) | reproduced· optimizedT1 | 2024-12-17Epoch AI |
| Fiction.LiveBench | 83.3accuracy (%) | unverifiedT1 | 2024-12-17Fiction.live leaderboard |
| GeoBench | 80accuracy (%) | unverifiedT1 | 2024-12-17GeoBench leaderboard |
| OTIS Mock AIME 2024-2025 | 73.31accuracy (%) | reproduced· optimizedT1 | 2024-12-17Epoch AI |
| Lech Mazur Writing | 70.2accuracy (%) | unverifiedT1 | 2024-12-17lechmazur/writing Github repository |
| GPQA diamond | 69.02accuracy (%) | reproduced· optimizedT1 | 2024-12-17Epoch AI |
| Aider polyglot | 61.7accuracy (%) | unverified· optimizedT1 | 2024-12-17https://aider.chat/docs/leaderboards/ |
| DTBench | 57.78accuracy (%) | unverifiedT1 | 2024-12-17https://conceptualreasoning.ai/dtbench |
| CadEval | 56accuracy (%) | unverifiedT1 | 2024-12-17CadEval Dashboard |
| METR Time Horizons | 51.05accuracy (%) | self-reportedT1 | 2024-12-17https://metr.org/blog/2025-03-19-measuring-ai-ability-to-complete-long-tasks/ |
| WeirdML | 46.08accuracy (%) | unverifiedT1 | 2024-12-17https://htihle.github.io/weirdml.html |
| SimpleQA Verified | 41.1accuracy (%) | unverifiedT2 | 2024-12-17 |
| ARC-AGI | 30.7accuracy (%) | unverified· optimizedT1 | 2024-12-17ARC Prize Leaderboard |
| SimpleBench | 28.12accuracy (%) | unverifiedT1 | 2024-12-17SimpleBench Leaderboard |
| LMCA | 26.2accuracy (%) | unverifiedT1 | 2024-12-17https://conceptualreasoning.ai/lmca |
| FrontierMath-Tiers-1-3-v2-Private | 14.74accuracy (%) | unverifiedT2 | 2024-12-17 |
| VPCT | 5.5accuracy (%) | unverifiedT1 | 2024-12-17VPCT leaderboard |
| Chess Puzzles | 5.3accuracy (%) | unverifiedT2 | 2024-12-17 |
| HLE | 3.32accuracy (%) | unverifiedT2 | 2024-12-17 |
Claims drawn from cited facts, not live model generation.
This model was developed by OpenAI and released in 2024.
2 cited facts
This model has scores on 19 tracked benchmarks and currently leads 0 of them. The record’s evidence status is independently reproduced.
3 cited facts
Its mean gap to the best recorded score across its tracked benchmarks is 38.81 points.
1 cited fact
6 facts cross-checked across data sources: 3 corroborated, 1 single-source, 2 disputed.
Source facts, citations, and refresh stamp for this record.
Sources: epoch_benchmarks, litellm_prices, modelsdev_models, openrouter_models
o1 is an AI model developed by OpenAI, released 2024-12-17. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.
o1 has recorded scores on 19 benchmarks, each shown with its evidence status.
o1 has recorded scores on 19 benchmarks — Lech Mazur Writing, GPQA diamond, MATH level 5, OTIS Mock AIME 2024-2025, DTBench, WeirdML, and 13 more. The full table above shows each score with its evidence status.
3 of 19 recorded scores (16%) are independently reproduced rather than self-reported by the lab.
o1 has 19 tracked claims: 3 independently reproduced, 1 self-reported, 15 unverified.
Listed API pricing: $15 per million input tokens, $60 per million output tokens. See the pricing block for the full breakdown.