Claude Opus 4.6 vs DeepSeek V3.2
Head to head on 26 shared benchmarks, with price, context window, and release dates.
Benchmark scores
46evals · sort by any column, ⤢ to expand fullscreen
| Model | |||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Claude Opus 4.6 Anthropic | 67.0% | 71.0% | 1804elo | 77.7 | 40.7% | 84.0% | 19.1% | 44.6% | 84.9% | 78.8% | 78.2% | 69.9% | 63.3% | 83.3% | 89.3% | 88.7% | 78.5% | 50.0% | 45.7% | 72.0% | 75.6% | 76.0% | 48.5% | 80.6 | 57.6% | 84.8% | - | - | - | 99.79 | 56.1% | 1.5 | - | 2.08 | 1223elo | - | 88771 | 58 | - | 48.2% | 86.7% | - | 68.5% | 93 | 51.6% | 44.5 | 66.5% |
| DeepSeek V3.2 DeepSeek | 47.9% | 16.0% | 1511elo | 66.8 | 22.1% | 75.1% | 11.2% | 49.0% | 68.0% | 65.9% | 64.6% | 50.0% | 48.2% | 70.4% | 85.0% | 77.2% | 57.5% | 8.0% | 38.7% | 59.0% | 70.0% | 68.2% | 32.6% | 65 | 5.1% | 78.9% | 74.7% | 74.2% | 59.0% | - | - | - | 0.0% | - | - | 31.3% | - | - | 59.3% | - | - | 83.7% | - | - | - | - | 51.7% |
2 / 2 models
| Attribute | Claude Opus 4.6 Anthropic | DeepSeek V3.2 DeepSeek |
|---|---|---|
| Description | Claude Opus 4.6 is an AI model from Anthropic. | DeepSeek V3.2 is an AI model from DeepSeek, released with open weights. |
| Family | Claude | Deepseek |
| Params | - | - |
| Open weights | closed | open |
| License | proprietary | mit |
| Released | 5 Feb 2026 | 1 Dec 2025 |
| Context | 1,000,000 | 163,840 |
| Input $/1M | $5.00 | $0.28 |
| Output $/1M | $25.00 | $0.42 |
| Cache read $/1M | $0.500 | $0.130 |
| Cache write $/1M | $6.250 | - |
| Speed | 0 tok/s | 0 tok/s |
| Latency | 0.00 s | 0.00 s |
| Modalities | text, image, file | text |
| Providers | - | - |
| Publisher | Anthropic | DeepSeek |
| 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) | Aider Polyglot Benchmark74.2% (Aider) MathArena57.47% GPQA Diamond75.1% (AA) Humanity's Last Exam (HLE)11.2% (AA) |
FAQ
- Is Claude Opus 4.6 or DeepSeek V3.2 better?
- Across the 26 benchmarks both models report on Sophon, Claude Opus 4.6 leads on 25 of them. Which one is "better" depends on the benchmark - the table above shows every shared score.
- How do Claude Opus 4.6 and DeepSeek V3.2 compare on GPQA Diamond?
- Claude Opus 4.6 scores 84.0% and DeepSeek V3.2 scores 75.1% on GPQA Diamond.
- Which is cheaper, Claude Opus 4.6 or DeepSeek V3.2?
- DeepSeek V3.2 is cheaper at $0.42 per million output tokens against $25.00 for Claude Opus 4.6 - about 59.5x.
- When were Claude Opus 4.6 and DeepSeek V3.2 released?
- Claude Opus 4.6 was released 5 Feb 2026 by Anthropic. DeepSeek V3.2 was released 1 Dec 2025 by DeepSeek.
Comparing something else? Build your own side-by-side.