Anthropic·released 2025-05-221 source
Claude Opus 4 benchmark scores: 22 benchmarks tracked. 27% of its scores are independently reproduced — the rest are self-reported or unverified.
Leads on 0 of 22 benchmarks·best result 85.05 on MATH level 5 (reproduced)·6 of 22 independently reproduced·$16.5/$82.5 per M tokens
Consensus: LiteLLM · See every model’s pricing →
| Benchmark | Score | Evidence status | Measured |
|---|---|---|---|
| MATH level 5 | 85.05accuracy (%) | reproduced· optimizedT1 | 2025-05-22Epoch AI |
| Lech Mazur Writing | 83.6accuracy (%) | unverifiedT1 | 2025-05-22lechmazur/writing Github repository |
| Aider polyglot | 72accuracy (%) | unverified· optimizedT1 | 2025-05-22https://aider.chat/docs/leaderboards/ |
| SWE-Bench verified | 70.66accuracy (%) | reproduced· optimizedT1 | 2025-05-22Epoch AI |
| DTBench | 69.33accuracy (%) | unverifiedT1 | 2025-05-22https://conceptualreasoning.ai/dtbench |
| GPQA diamond | 68.35accuracy (%) | reproduced· optimizedT1 | 2025-05-22Epoch AI |
| OTIS Mock AIME 2024-2025 | 64.41accuracy (%) | reproduced· optimizedT1 | 2025-05-22Epoch AI |
| METR Time Horizons | 63.93accuracy (%) | unverifiedT1 | 2025-05-22METR - Measuring AI Ability to Complete Long Tasks |
| Fiction.LiveBench | 61.1accuracy (%) | unverifiedT1 | 2025-05-22Fiction.live leaderboard |
| SimpleBench | 50.56accuracy (%) | unverifiedT1 | 2025-05-22SimpleBench Leaderboard |
| GeoBench | 49accuracy (%) | unverifiedT1 | 2025-05-22GeoBench leaderboard |
| DeepResearch Bench | 46.8accuracy (%) | unverifiedT1 | 2025-05-22https://evals.futuresearch.ai/#drb |
| LMCA | 44.04accuracy (%) | unverifiedT1 | 2025-05-22https://conceptualreasoning.ai/lmca |
| WeirdML | 43.72accuracy (%) | unverifiedT1 | 2025-05-22https://htihle.github.io/weirdml.html |
| Cybench | 38accuracy (%) | unverified· optimizedT1 | 2025-05-22Cybench leaderboard |
| ARC-AGI | 35.67accuracy (%) | unverified· optimizedT1 | 2025-05-22https://arcprize.org/leaderboard |
| ARC-AGI-2 | 8.61accuracy (%) | unverifiedT2 | 2025-05-22 |
| FrontierMath-2025-02-28-Private | 7.86accuracy (%) | reproduced· optimizedT1 | 2025-05-22Epoch AI |
| VPCT | 7accuracy (%) | unverifiedT1 | 2025-05-22VPCT leaderboard |
| FrontierMath-Tier-4-2025-07-01-Private | 6.94accuracy (%) | reproducedT1 | 2025-05-22Epoch AI |
| GSO-Bench | 6.9accuracy (%) | unverifiedT1 | 2025-05-22GSO Leaderboard |
| HLE | 6.22accuracy (%) | unverifiedT2 | 2025-05-22 |
Claims drawn from cited facts, not live model generation.
Developed by Anthropic, this model was released in 2025.
2 cited facts
This model is scored on 22 tracked benchmarks. It holds no current top score among them. Its results have been independently reproduced, meaning outside evaluation backs the scores up.
3 cited facts
This model trails the top benchmark score by an average of 35.73 points, a wide score gap under the disclosed harness that reads as well behind the leaders rather than at the frontier. It also holds the top score on 0 benchmarks, meaning it has no category leadership yet.
2 cited facts
5 facts cross-checked across data sources: 5 single-source.
Source facts, citations, and refresh stamp for this record.
Sources: epoch_benchmarks, litellm_prices
Claude Opus 4 is an AI model developed by Anthropic, released 2025-05-22. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.
Claude Opus 4 has recorded scores on 22 benchmarks, each shown with its evidence status.
Claude Opus 4 has recorded scores on 22 benchmarks — Lech Mazur Writing, GPQA diamond, MATH level 5, OTIS Mock AIME 2024-2025, DTBench, WeirdML, and 16 more. The full table above shows each score with its evidence status.
6 of 22 recorded scores (27%) are independently reproduced rather than self-reported by the lab.
Claude Opus 4 has 22 tracked claims: 6 independently reproduced, 16 unverified.
Listed API pricing: $16.5 per million input tokens, $82.5 per million output tokens. See the pricing block for the full breakdown.