Updated daily at 05:00 UTC
Vision-language
Claude Opus 4.6
- Released
- 5 Feb 2026
- Status
- Legacy
- Context
- 1M
- Max output
- 128K
- Input per 1M tokens
- $5.00
- Output per 1M tokens
- $25.00
- Open weights
- No
- License
- proprietary
- Parameters
- —
- Knowledge cutoff
- May 2025
- Reasoning
- Yes
- Modalities
- text, image → text
Category scores
Out of 100, from independent and third-party results only, by method category-2026-09-29.1.
| Category | Score | Rank | Coverage |
|---|---|---|---|
| Knowledge & reasoning | 38.0 | 39 | 4 of 4 |
| Coding & software engineering | 33.9 | Not ranked | 2 of 5 |
| Agentic tool use | 55.3 | Not ranked | 1 of 2 |
| Vision & multimodal understanding | 0.0 | Not ranked | 1 of 3 |
| Factuality & hallucination | 47.0 | 29 | 1 of 1 |
| Human preference | 58.5 | 8 | 4 of 5 |
Benchmark results
Every result we hold for this model, with who ran it and where it was published.
Knowledge & reasoning
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| Chess Puzzles | Puzzles solved | 17.0% | Independent | 32K | 6 Feb 2026 | Epoch AI |
| Chess Puzzles | Puzzles solved | 14.0% | Independent | max | 6 Aug 2026 | Epoch AI |
| Chess Puzzles | Puzzles solved | 13.0% | Independent | 120K | 20 Feb 2026 | Epoch AI |
| Chess Puzzles | Puzzles solved | 10.0% | Independent | 64K | 6 Feb 2026 | Epoch AI |
| DTBench (Decision Theory Benchmark) | Accuracy (capability questions) | 91.2% | Third party | max | — | https://conceptualreasoning.ai/dtbench via Epoch AI |
| Earthborne Rangers (EBR-bench) | Objectives reached | 12.7% | Independent | max | 29 Jun 2026 | Epoch AI |
| LMCA (Language Model Conceptual Argumentation) | Correlation with expert ratings (Pearson r) | 0.5576 | Third party | max | — | https://conceptualreasoning.ai/lmca via Epoch AI |
| Mystery Game Puzzles | Puzzles solved | 25.0% | Independent | max | 25 Jul 2026 | Epoch AI |
| Mystery Game Puzzles | Puzzles solved | 15.0% | Independent | — | 6 Aug 2026 | Epoch AI |
| Mystery Game Puzzles | Puzzles solved | 7.0% | Independent | low | 6 Aug 2026 | Epoch AI |
| SimpleBench | Score (AVG@5) | 67.6% | Third party | — | — | https://lmcouncil.ai/benchmarks via Epoch AI |
Scientific reasoning
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| GPQA Diamond | Accuracy | 90.5% | Independent | 32K | 6 Feb 2026 | Epoch AI |
| GPQA Diamond | Accuracy | 88.8% | Independent | 64K | 6 Feb 2026 | Epoch AI |
| GPQA Diamond | Accuracy | 88.4% | Independent | max | 6 Aug 2026 | Epoch AI |
Mathematics
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| FrontierMath Tier 4 | Accuracy | 26.8% | Independent | max | 11 Jun 2026 | Epoch AI |
| FrontierMath Tier 4 (v1) | Accuracy | 22.9% | Independent | max | 12 Feb 2026 | Epoch AI |
| FrontierMath Tier 4 (v1) | Accuracy | 20.8% | Independent | 64K | 6 Feb 2026 | Epoch AI |
| FrontierMath Tier 4 (v1) | Accuracy | 20.8% | Independent | 32K | 6 Feb 2026 | Epoch AI |
| FrontierMath Tier 4 (v1) | Accuracy | 14.6% | Independent | — | 5 Feb 2026 | Epoch AI |
| FrontierMath Tiers 1-3 | Accuracy | 66.0% | Independent | max | 11 Jun 2026 | Epoch AI |
| FrontierMath Tiers 1-3 (v1) | Accuracy | 40.7% | Independent | max | 12 Feb 2026 | Epoch AI |
| FrontierMath Tiers 1-3 (v1) | Accuracy | 40.0% | Independent | 32K | 6 Feb 2026 | Epoch AI |
| FrontierMath Tiers 1-3 (v1) | Accuracy | 39.7% | Independent | 64K | 6 Feb 2026 | Epoch AI |
| FrontierMath Tiers 1-3 (v1) | Accuracy | 38.3% | Independent | — | 5 Feb 2026 | Epoch AI |
| OTIS Mock AIME 2024-2025 | Accuracy | 94.4% | Independent | 64K | 6 Feb 2026 | Epoch AI |
| OTIS Mock AIME 2024-2025 | Accuracy | 93.1% | Independent | 32K | 6 Feb 2026 | Epoch AI |
| OTIS Mock AIME 2024-2025 | Accuracy | 91.1% | Independent | max | 6 Aug 2026 | Epoch AI |
Coding & software engineering
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| Cybench | Unguided challenges solved | 93.0% | Self-reported | unknown | — | Opus 4.6 System Card via Epoch AI |
| FrontierCode | Main score (Mean@5) | 26.6% | Third party | unknown | — | https://cognition.com/frontiercode via Epoch AI |
| GSO | OPT@1 | 41.2% | Third party | high | — | GSO Leaderboard via Epoch AI |
| GSO | OPT@1 | 33.3% | Third party | unknown | — | GSO Leaderboard via Epoch AI |
| GSO | OPT@1 | 33.3% | Third party | — | — | GSO Leaderboard via Epoch AI |
| METR Time Horizons | Average task success (Epoch's column) | 78.9% | Source not stated | unknown | — | Epoch AI via Epoch AI |
| SWE-bench Verified | % resolved | 78.7% | Independent | — | 18 Feb 2026 | Epoch AI |
| SWE-bench Verified | % resolved | 75.6% | Independent | — | 10 Feb 2026 | Epoch AI |
| Terminal-Bench 2.0 | Accuracy (mean over trials) | 79.8% | Third party | unknown | — | https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI |
| Terminal-Bench 2.0 | Accuracy (mean over trials) | 76.4% | Third party | unknown | — | https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI |
| Terminal-Bench 2.0 | Accuracy (mean over trials) | 75.3% | Third party | unknown | — | https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI |
| Terminal-Bench 2.0 | Accuracy (mean over trials) | 74.7% | Third party | unknown | — | https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI |
| Terminal-Bench 2.0 | Accuracy (mean over trials) | 72.1% | Third party | unknown | — | https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI |
| Terminal-Bench 2.0 | Accuracy (mean over trials) | 71.9% | Third party | unknown | — | https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI |
| Terminal-Bench 2.0 | Accuracy (mean over trials) | 69.9% | Third party | — | — | Terminal-Bench v2 Leaderboard via Epoch AI |
| Terminal-Bench 2.0 | Accuracy (mean over trials) | 69.9% | Third party | unknown | — | https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI |
| Terminal-Bench 2.0 | Accuracy (mean over trials) | 66.9% | Third party | unknown | — | https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI |
| Terminal-Bench 2.0 | Accuracy (mean over trials) | 66.5% | Third party | unknown | — | https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI |
| Terminal-Bench 2.0 | Accuracy (mean over trials) | 62.9% | Third party | unknown | — | https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI |
| Terminal-Bench 2.0 | Accuracy (mean over trials) | 58.0% | Third party | unknown | — | https://www.tbench.ai/leaderboard/terminal-bench/2.0 via Epoch AI |
| WeirdML (v2) | Average accuracy | 78.0% | Third party | high | — | https://htihle.github.io/weirdml.html via Epoch AI |
| WeirdML (v2) | Average accuracy | 77.9% | Third party | unknown | — | WeirdML Leaderboard via Epoch AI |
| WeirdML (v2) | Average accuracy | 65.9% | Third party | — | — | https://htihle.github.io/weirdml.html via Epoch AI |
Agentic tool use
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| Deep Research Bench (FutureSearch) | Average score | 55.3% | Third party | high | — | https://evals.futuresearch.ai/#drb via Epoch AI |
| Deep Research Bench (FutureSearch) | Average score | 53.2% | Third party | medium | — | https://evals.futuresearch.ai/#drb via Epoch AI |
| Deep Research Bench (FutureSearch) | Average score | 51.4% | Third party | low | — | https://evals.futuresearch.ai/#drb via Epoch AI |
Professional knowledge work
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| APEX-Agents | Pass@1 (all rubric criteria met) | 46.3% | Source not stated | max | — | Epoch AI via Epoch AI |
Long context
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| CL-bench | Tasks solved | 20.7% | Source not stated | unknown | — | Epoch AI via Epoch AI |
| CL-bench Life | Tasks solved | 17.0% | Source not stated | high | — | Epoch AI via Epoch AI |
| CL-bench Life | Tasks solved | 13.6% | Source not stated | unknown | — | Epoch AI via Epoch AI |
Vision & multimodal understanding
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| Furniture Assembly | Score | 28.3% | Independent | max | 10 Sep 2026 | Epoch AI |
Factuality & hallucination
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| SimpleQA Verified | Accuracy (% correct) | 47.0% | Independent | max | 27 Aug 2026 | Epoch AI |
Human preference
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| Code Arena: WebDev (arena.ai) | Arena score (Elo scale) | 1546 | Independent | high | 25 Sep 2026 | Arena |
| Code Arena: WebDev (arena.ai) | Arena score (Elo scale) | 1537 | Independent | — | 25 Sep 2026 | Arena |
| Code Arena: WebDev (arena.ai) | Votes | 20788 | Independent | — | 25 Sep 2026 | Arena |
| Code Arena: WebDev (arena.ai) | Votes | 19527 | Independent | high | 25 Sep 2026 | Arena |
| Document Arena (arena.ai) | Arena score (Elo scale) | 1495 | Independent | high | 13 Sep 2026 | Arena |
| Document Arena (arena.ai) | Arena score (Elo scale) | 1494 | Independent | — | 13 Sep 2026 | Arena |
| Document Arena (arena.ai) | Votes | 41260 | Independent | — | 13 Sep 2026 | Arena |
| Document Arena (arena.ai) | Votes | 27929 | Independent | high | 13 Sep 2026 | Arena |
| Text Arena (arena.ai) | Arena score (Elo scale) | 1505 | Independent | high | 25 Sep 2026 | Arena |
| Text Arena (arena.ai) | Arena score (Elo scale) | 1498 | Independent | — | 25 Sep 2026 | Arena |
| Text Arena (arena.ai) | Votes | 80836 | Independent | — | 25 Sep 2026 | Arena |
| Text Arena (arena.ai) | Votes | 76518 | Independent | high | 25 Sep 2026 | Arena |
| Vision Arena (arena.ai) | Arena score (Elo scale) | 1299 | Independent | high | 28 Sep 2026 | Arena |
| Vision Arena (arena.ai) | Arena score (Elo scale) | 1295 | Independent | — | 28 Sep 2026 | Arena |
| Vision Arena (arena.ai) | Votes | 26904 | Independent | — | 28 Sep 2026 | Arena |
| Vision Arena (arena.ai) | Votes | 22350 | Independent | high | 28 Sep 2026 | Arena |
Other Claude Opus versions
- Claude Opus 5.522 Sep 2026
- Claude Opus 524 Jul 2026
- Claude Opus 4.828 May 2026
- Claude Opus 4.716 Apr 2026
- Claude Opus 4.524 Nov 2025
- Claude Opus 4.15 Aug 2025
- Claude Opus 422 May 2025
- Claude 3 Opus4 Mar 2024