Claude Opus 4.6 vs Claude Opus 4.8
Head to head on 28 shared benchmarks, with price, context window, and release dates.
Benchmark scores
53evals · sort by any column, ⤢ to expand fullscreen
| Model | ||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Claude Opus 4.6 Anthropic | 56.1% | 67.0% | 71.0% | 1223elo | 1804elo | 77.7 | 40.7% | 84.0% | 19.1% | 44.6% | 78.8% | 78.2% | 69.9% | 63.3% | 83.3% | 89.3% | 88.7% | 78.5% | 48.2% | 86.7% | 68.5% | 50.0% | 51.6% | 45.7% | 76.0% | 48.5% | 57.6% | 84.8% | 99.79 | 1.5 | - | - | 2.08 | - | - | - | - | - | - | 84.9% | 88771 | 58 | - | 93 | - | 44.5 | 72.0% | 75.6% | - | - | 80.6 | - | - | 66.5% |
| Claude Opus 4.8 Anthropic | 15.5% | 66.7% | 67.0% | 1281elo | 1835elo | 80.8 | 47.2% | 92.0% | 48.7% | 62.2% | 80.1% | 79.3% | 78.3% | 67.4% | 81.4% | 84.3% | 89.7% | 91.8% | 53.2% | 85.8% | 69.9% | 69.0% | 54.8% | 53.5% | 75.6% | 58.3% | 82.7% | 94.4% | - | - | 80.4% | 14.5% | - | 40.0% | 53.9% | 13.4% | 22.5% | 1890elo | 56.9% | - | - | - | 10.4% | - | 83.4% | - | - | - | 69.2% | 82.7% | - | 60.9% | 70.9% | 63.4% |
2 / 2 models
| Attribute | Claude Opus 4.6 Anthropic | Claude Opus 4.8 Anthropic |
|---|---|---|
| Description | Claude Opus 4.6 is an AI model from Anthropic. | Claude Opus 4.8 (Adaptive Reasoning, Max Effort) is an AI model from Anthropic. |
| Family | Claude | Claude |
| Params | - | - |
| Open weights | closed | - |
| License | proprietary | - |
| Released | 5 Feb 2026 | 28 May 2026 |
| Context | 1,000,000 | 1,000,000 |
| Input $/1M | $5.00 | $5.00 |
| Output $/1M | $25.00 | $25.00 |
| Cache read $/1M | $0.500 | $0.500 |
| Cache write $/1M | $6.250 | $6.250 |
| Speed | 0 tok/s | 0 tok/s |
| Latency | 0.00 s | 0.00 s |
| Modalities | text, image, file | text, image, file |
| Providers | - | - |
| Publisher | Anthropic | Anthropic |
| Reported scores | GPQA Diamond84.0% (AA) Humanity's Last Exam (HLE)19.1% (AA) SciCode45.7% (AA) τ²-bench (Tau²-bench)84.8% (AA) Terminal-Bench (Hard)48.5% (AA) |
FAQ
- Is Claude Opus 4.6 or Claude Opus 4.8 better?
- Across the 28 benchmarks both models report on Sophon, Claude Opus 4.8 leads on 21 of them. Which one is "better" depends on the benchmark - the table above shows every shared score.
- How do Claude Opus 4.6 and Claude Opus 4.8 compare on GPQA Diamond?
- Claude Opus 4.6 scores 84.0% and Claude Opus 4.8 scores 92.0% on GPQA Diamond.
- When were Claude Opus 4.6 and Claude Opus 4.8 released?
- Claude Opus 4.6 was released 5 Feb 2026 by Anthropic. Claude Opus 4.8 was released 28 May 2026 by Anthropic.
Comparing something else? Build your own side-by-side.