Claude Opus 4.7 vs GPT-5.2
Head to head on 26 shared benchmarks, with price, context window, and release dates.
Benchmark scores
52evals · sort by any column, ⤢ to expand fullscreen
| Model | |||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Claude Opus 4.7 Anthropic | 66.1% | 79.0% | 1906elo | 81.8 | 43.8% | 88.5% | 33.3% | 43.6% | 79.7% | 82.1% | 78.3% | 59.3% | 77.9% | 93.1% | 87.7% | 73.2% | 54.9% | 83.0% | 70.3% | 54.0% | 56.1% | 50.1% | 75.3% | 54.5% | 71.0% | 74.0% | - | - | 3.21 | - | - | - | - | - | 1311elo | 51.5% | - | - | - | - | - | - | 91.5 | - | 45.3% | 100.0% | - | - | - | 72.8 | 56.1% | 67.4% | 67.2% |
| GPT-5.2 OpenAI | 65.9% | 60.0% | 1699elo | 71.6 | 40.7% | 71.2% | 8.0% | 47.4% | 78.7% | 76.1% | 78.2% | 61.8% | 79.8% | 93.2% | 83.2% | 72.0% | 49.7% | 84.4% | 67.1% | 15.0% | 49.3% | 40.4% | 75.8% | 31.8% | 53.5% | 46.5% | 51.0% | 100 | - | 100.0% | 0.0% | 16.7% | 40.6% | 9.4% | - | - | 12.93 | 85.0% | 79.0% | 66.9% | 24.0% | 81.4% | - | 96.7 | - | - | 18.7% | 66.7% | 71.8% | - | - | - | 56.3% |
2 / 2 models
| Attribute | Claude Opus 4.7 Anthropic | GPT-5.2 OpenAI |
|---|---|---|
| Description | Claude Opus 4.7 is an AI model from Anthropic. | GPT-5.2 is an AI model from OpenAI. |
| Family | Claude | GPT |
| Params | - | - |
| Open weights | closed | closed |
| License | proprietary | proprietary |
| Released | 16 Apr 2026 | 11 Dec 2025 |
| Context | 1,000,000 | 400,000 |
| Input $/1M | $5.00 | $1.75 |
| Output $/1M | $25.00 | $14.00 |
| Cache read $/1M | $0.500 | $0.175 |
| Cache write $/1M | $6.250 | - |
| Speed | 0 tok/s | 0 tok/s |
| Latency | 0.00 s | 0.00 s |
| Modalities | text, image, file | file, image, text |
| Providers | - | - |
| Publisher | Anthropic | OpenAI |
| Reported scores | GPQA Diamond88.5% (AA) Humanity's Last Exam (HLE)33.3% (AA) IFBench43.6% (AA) τ²-bench (Tau²-bench)74.0% (AA) Terminal-Bench (Hard)54.5% (AA) |
FAQ
- Is Claude Opus 4.7 or GPT-5.2 better?
- Across the 26 benchmarks both models report on Sophon, Claude Opus 4.7 leads on 20 of them. Which one is "better" depends on the benchmark - the table above shows every shared score.
- How do Claude Opus 4.7 and GPT-5.2 compare on GPQA Diamond?
- Claude Opus 4.7 scores 88.5% and GPT-5.2 scores 71.2% on GPQA Diamond.
- Which is cheaper, Claude Opus 4.7 or GPT-5.2?
- GPT-5.2 is cheaper at $14.00 per million output tokens against $25.00 for Claude Opus 4.7 - about 1.8x.
- When were Claude Opus 4.7 and GPT-5.2 released?
- Claude Opus 4.7 was released 16 Apr 2026 by Anthropic. GPT-5.2 was released 11 Dec 2025 by OpenAI.
Comparing something else? Build your own side-by-side.