Claude Opus 4.8 vs Claude Sonnet 4.6
Head to head on 29 shared benchmarks, with price, context window, and release dates.
Benchmark scores
48evals · sort by any column, ⤢ to expand fullscreen
| Model | |||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Claude Opus 4.8 Anthropic | 15.5% | 66.7% | 67.0% | 1281elo | 1835elo | 80.8 | 53.9% | 47.2% | 92.0% | 48.7% | 62.2% | 80.1% | 79.3% | 78.3% | 67.4% | 81.4% | 84.3% | 89.7% | 69.9% | 83.4% | 69.0% | 54.8% | 53.5% | 75.6% | 58.3% | 60.9% | 70.9% | 82.7% | 94.4% | - | - | 80.4% | 14.5% | - | 40.0% | 13.4% | 22.5% | 1890elo | 56.9% | - | 10.4% | 91.8% | 53.2% | 85.8% | - | 69.2% | 82.7% | - | 63.4% |
| Claude Sonnet 4.6 Anthropic | 47.6% | 65.3% | 78.0% | 1207elo | 1805elo | 79.9 | 51.0% | 32.4% | 79.7% | 11.2% | 42.4% | 78.4% | 80.0% | 76.1% | 63.9% | 77.7% | 86.5% | 86.4% | 67.7% | 72.1% | 45.0% | 46.6% | 44.1% | 77.1% | 42.4% | 50.6% | 60.6% | 51.5% | 78.9% | 87.1% | 47.2% | - | - | 1.35 | - | - | - | - | - | 92.2% | - | - | - | - | 89.3 | - | - | 76.3 | 62.8% |
2 / 2 models
| Attribute | Claude Opus 4.8 Anthropic | Claude Sonnet 4.6 Anthropic |
|---|---|---|
| Description | Claude Opus 4.8 (Adaptive Reasoning, Max Effort) is an AI model from Anthropic. | Claude Sonnet 4.6 is an AI model from Anthropic. |
| Family | Claude | Claude |
| Params | - | - |
| Open weights | - | closed |
| License | - | proprietary |
| Released | 28 May 2026 | 17 Feb 2026 |
| Context | 1,000,000 | 1,000,000 |
| Input $/1M | $5.00 | $3.00 |
| Output $/1M | $25.00 | $15.00 |
| Cache read $/1M | $0.500 | $0.300 |
| Cache write $/1M | $6.250 | $3.750 |
| Speed | 0 tok/s | 45 tok/s |
| Latency | 0.00 s | 1.12 s |
| Modalities | text, image, file | text, image, file |
| Providers | - | - |
| Publisher | Anthropic | Anthropic |
| Reported scores | GPQA Diamond79.7% (AA) Humanity's Last Exam (HLE)11.2% (AA) IFBench42.4% (AA) SciCode44.1% (AA) τ²-bench (Tau²-bench)78.9% (AA) |
FAQ
- Is Claude Opus 4.8 or Claude Sonnet 4.6 better?
- Across the 29 benchmarks both models report on Sophon, Claude Opus 4.8 leads on 24 of them. Which one is "better" depends on the benchmark - the table above shows every shared score.
- How do Claude Opus 4.8 and Claude Sonnet 4.6 compare on GPQA Diamond?
- Claude Opus 4.8 scores 92.0% and Claude Sonnet 4.6 scores 79.7% on GPQA Diamond.
- Which is cheaper, Claude Opus 4.8 or Claude Sonnet 4.6?
- Claude Sonnet 4.6 is cheaper at $15.00 per million output tokens against $25.00 for Claude Opus 4.8 - about 1.7x.
- When were Claude Opus 4.8 and Claude Sonnet 4.6 released?
- Claude Opus 4.8 was released 28 May 2026 by Anthropic. Claude Sonnet 4.6 was released 17 Feb 2026 by Anthropic.
Comparing something else? Build your own side-by-side.