Claude Opus 4
Claude Opus 4 (Anthropic) was released on May 22, 2025. In the Epoch AI capability ranking (ECI), it ranks #108 out of 267 models, with 142.65. On Arena, which measures human preference in blind comparisons, it has 1,442 ELO points (#113). Against models of the same generation, its best result is on ARC-AGI 2: 8.6%, #3 of 26.
Where Claude Opus 4 stands out and where it falls behind, by the numbers
Claude Opus 4 has published results on 26 tests compiled by Epoch AI. The yardstick here is its generation: its position among the models released between Aug 25, 2024 and Aug 20, 2025 that took the same test. Against the whole archive, which includes 2023 models, a recent release lands near the top of almost everything, and that says nothing.
Where it stands out
Tests where it sits in the top third of its own generation.
Where it falls behind
Tests where it sits in the bottom half of its own generation.
See all 26 tests, one by one
| Test | Score | In its generation | Overall | Test record |
|---|---|---|---|---|
| GPQA Diamond | 76.3% | #13 / 60 (tied with 1) | #93 / 173 | 96.4% Gemini 4 Argon |
| AIME 2024 and 2025 (OTIS) | 64.4% | #20 / 58 | #106 / 168 | 100.0% GPT-5.5 |
| Dtbench | 81.6% | #5 / 46 | #67 / 148 | 98.9% Claude Opus 5.5 |
| Lmca | 37.43 | #4 / 36 | #57 / 121 | 68.23 Claude Opus 5.5 |
| CritPt | 0.3% | #8 / 26 | #83 / 112 | 32.3% GPT-5.6 Sol |
| WeirdML | 43.7% | #12 / 37 | #61 / 112 | 93.6% GPT-6 Astra |
| ARC-AGI 2 | 8.6% | #3 / 26 | #51 / 83 | 95.0% GPT-6 Astra |
| ARC-AGI | 35.7% | #9 / 25 | #63 / 81 | 98.5% Claude Fable 5 |
| SimpleBench | 58.8% | #4 / 27 | #27 / 74 | 81.9% Claude Fable 5 |
| MATH level 5 | 85.0% | #16 / 43 | #19 / 73 | 98.1% GPT-5 |
| FrontierMath | 4.5% | #14 / 31 (tied with 1) | #46 / 65 | 51.7% GPT-5.5 |
| ForecastBench | 61.10 | #8 / 23 | #14 / 62 | 62.50 o3 |
| FrontierMath tier 4 | 4.2% | #4 / 15 (tied with 3) | #23 / 48 | 39.6% GPT-5.5 Pro |
| Hle | 10.7% | #7 / 18 | #26 / 39 | 54.8% GPT-6 Astra |
| Aider Polyglot | 72.00 | #6 / 32 (tied with 1) | #8 / 38 | 88.00 GPT-5 |
| Creative writing (Lech Mazur) | 8.36 | #6 / 34 | #7 / 38 | 8.60 GPT-5 |
| EnigmaEval | 5.6% | #7 / 17 (tied with 1) | #21 / 36 | 39.3% Claude Fable 5 |
| SWE-bench Verified | 70.7% | #3 / 8 | #24 / 32 | 83.5% Claude Opus 4.7 |
| Fiction.liveBench | 37.5% | #19 / 27 (tied with 1) | #23 / 31 | 100.0% o3 |
| METR: time horizon | 100 | #5 / 15 | #13 / 29 | 719 Claude Opus 4.6 |
| Gso | 6.9% | #2 / 9 | #19 / 28 | 88.2% Claude Fable 5.1 |
| DeepResearch Bench | 46.8% | #4 / 9 | #12 / 24 | 55.3% Claude Opus 4.6 |
| GeoBench | 2,915 | #14 / 18 | #20 / 24 | 4,333 Gemini 3 Flash |
| VPCT | 38.0% | #8 / 13 | #16 / 22 | 91.0% Gemini 3 Pro |
| Cybench | 38.0% | #3 / 10 | #6 / 19 | 93.0% Claude Opus 4.6 |
| Algotune | 1.33 | #8 / 8 | #15 / 15 | 2.05 GPT-5.2 |
For each test, the model's best effort level counts, as in the ranking. Source of the scores: Epoch AI, under CC BY 4.0.
What to compare Claude Opus 4 with
The models from other labs closest in capability and released in the same period, at most two per lab. Click any of them to see its page.
| Model | Lab | ECI | ELO | LiveBench | Release |
|---|---|---|---|---|---|
| GLM-4.7 | Z.ai (Zhipu AI) | 143.79 | 1,495 | n/a | Dec 22, 2025 |
| Gemini 2.5 Flash (Sep 2025) | Google DeepMind | 143.00 | 1,450 | n/a | Sep 25, 2025 |
| Claude Opus 4 | Anthropic | 142.65 | 1,442 | n/a | May 22, 2025 |
| Qwen3-Max-Instruct | Alibaba | 142.46 | n/a | n/a | Sep 24, 2025 |
| Qwen3-Max | Alibaba | 142.34 | 1,478 | n/a | Sep 24, 2025 |
| o1 | OpenAI | 141.84 | 1,383 | n/a | Dec 17, 2024 |
| DeepSeek-R1 (May 2025) | DeepSeek | 141.27 | 1,400 | n/a | May 28, 2025 |
Claude Opus 4 or Qwen3-Max?
Qwen3-Max is the model from another lab closest to Claude Opus 4 in capability. Across the 5 tests both took, this is the score:
| Test | Claude Opus 4 | Qwen3-Max |
|---|---|---|
| Lmca | 37.43 | 28.27 |
| MATH level 5 | 85.0% | 97.1% |
| AIME 2024 and 2025 (OTIS) | 64.4% | 73.3% |
| GPQA Diamond | 76.3% | 72.6% |
| Dtbench | 81.6% | 82.1% |
Unweighted score: each test counts one point, whether math or coding. The dot marks who did better on each row.
What Claude Opus 4 costs to deliver each task
Price per token is what is on the price sheet; cost per task is what lands on the invoice, because it depends on how many tokens the model spends. Each point is an effort level, published by whoever ran the test and compiled by Epoch AI. The further up and to the left, the more score per dollar.
ARC-AGI 2
The score scale does not start at zero.
ARC-AGI
The score scale does not start at zero.
Other models from Anthropic
The models from the same lab closest in time, newest first.
| Model | Release | ECI | ELO | LiveBench | Access |
|---|---|---|---|---|---|
| Claude Opus 4.5 | Nov 24, 2025 | 150.17 | 1,504 | 72.6 | API access |
| Claude Haiku 4.5 | Oct 15, 2025 | 142.40 | 1,454 | n/a | API access |
| Claude Sonnet 4.5 | Sep 29, 2025 | 146.83 | 1,488 | n/a | API access |
| Claude Opus 4.1 | Aug 5, 2025 | 144.09 | 1,479 | n/a | API access |
| Claude Opus 4 | May 22, 2025 | 142.65 | 1,442 | n/a | API access |
| Claude Sonnet 4 | May 22, 2025 | 141.67 | 1,414 | n/a | API access |
| Claude 3.7 Sonnet | Feb 24, 2025 | 141.14 | 1,360 | n/a | API access |
| Claude 3.5 Sonnet (October 2024) | Oct 22, 2024 | 133.53 | 1,342 | n/a | API access |
| Claude 3.5 Haiku | Oct 22, 2024 | 127.17 | 1,286 | n/a | API access |
Have you used Claude Opus 4? Rate it
Users' opinion, not measurement: the rating is kept apart from the ranking and does not move anyone's position. One vote per browser, no sign-up, and you can change your rating later.
What people ask about Claude Opus 4
Where does Claude Opus 4 rank among AI models?
In the ROO3 AI Benchmark, ordered by the Epoch AI ECI, Claude Opus 4 ranks #108 out of 267 models, with an ECI of 142.65. The ECI is a composite index that combines several hard benchmarks into a single scale. The position changes when a new model comes out, and the page is refreshed six times a day.
Is Claude Opus 4 better than Qwen3-Max?
Across the 5 tests both models took and Epoch AI compiled, Claude Opus 4 comes out ahead on 2, behind on 3 and tied on 0. It is a test-by-test count with no weighting: a win in math counts the same as a win in coding. To choose between the two, look at the tests that resemble your own use.
What is Claude Opus 4 best at?
Against models of the same generation (released up to 9 months before and up to 3 months after it), the best relative results of Claude Opus 4 are: ARC-AGI 2 (#3 of 26); Lmca (#4 of 36); Dtbench (#5 of 46). The yardstick is its position among the models of that period that took the same test.
Where do the numbers on this page come from?
From the same three sources as the ranking, all openly licensed: Epoch AI (ECI and individual tests, CC BY 4.0), Arena (human-preference ELO, CC BY 4.0) and LiveBench (overall average, CC BY-SA 4.0). ROO3 does not measure any model. When the page shows a figure published by the maker itself, it appears in the written analysis, labeled as the maker's, and never enters the ranking.
Where each number comes from
Epoch AI, "AI Benchmarking Hub". Published at epoch.ai. Data under the CC BY 4.0 licence, reordered by roo3.co.
The Arena (LMArena) leaderboard, from the public lmarena-ai/leaderboard-dataset on Hugging Face, under the CC BY 4.0 licence. We are not affiliated with the Arena.
LiveBench, by the LiveBench team, published at livebench.ai under the CC BY-SA 4.0 license. The overall average is recomputed by roo3.co with the same formula as the website (average of the categories), and those numbers stay under CC BY-SA 4.0. We are not affiliated with LiveBench.
← See the full ranking