Claude 4 Sonnet vs GPT-5
Head to head on 25 shared benchmarks, with price, context window, and release dates.
Benchmark scores
77evals · sort by any column, ⤢ to expand fullscreen
| Model | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| GPT-5 OpenAI | 94.6% | 48.3% | 68.1% | 61.1% | 57.7% | 1623elo | 72.8 | 83.3% | 68.6% | 6.6% | 45.0% | 54.3% | 86.1% | 49.6% | 83.7% | 93.3% | 82.0% | 65.5% | 43.7% | 37.8% | 74.4% | 73.4% | 12.9% | 0.86 | 0.0% | 46.7% | 91.1% | 88.0% | 0.0% | 0.15 | - | 0.0% | 4.58 | 80.0% | 80.0% | 91.7% | 32.4% | 0.0% | 88.9% | 100.0% | 48.9% | 83.4% | 58.2% | - | - | - | - | - | - | - | 37.5% | 0.52 | 1.62 | 100.0% | 0.0% | 92.7% | 92.7% | 10.0% | 0.0% | 60.0% | 0.26 | - | 0.0% | 18.0% | - | - | 74.9% | - | - | 2.34 | 87.5% | 20.1% | 54.8% | - | - | 90.0% | 90.0% | 56.7% |
| Claude 4 Sonnet Anthropic | 40.7% | 38.0% | 47.8% | 54.7% | 19.8% | 1480elo | 58.4 | 67.5% | 68.3% | 4.3% | 45.4% | 44.9% | 93.4% | 33.9% | 72.4% | 86.7% | 83.7% | 62.5% | 35.0% | 37.3% | 74.6% | 69.6% | 27.3% | 0.59 | 52.3% | - | - | - | - | - | 22.9% | - | - | - | - | - | - | - | - | - | - | - | - | 64.8% | 77.5% | 54.6% | 44.3% | 72.9% | 70.5% | 69.0% | - | - | - | - | - | - | - | - | - | - | - | 43.9% | - | - | 6.18 | 6.18 | - | 58.3% | 35.6% | - | - | - | - | 84.0% | 84.0% | - | - | 55.5% |
2 / 2 models
| Attribute | Claude 4 Sonnet Anthropic | GPT-5 OpenAI |
|---|---|---|
| Description | Claude 4 Sonnet is an AI model from Anthropic. | OpenAI's August 2025 unified frontier model that auto-routes between a fast model and a deeper "thinking" variant. |
| Family | Claude | GPT 5 |
| Params | - | Undisclosed (closed) |
| Open weights | closed | closed |
| License | proprietary | Proprietary |
| Released | 22 May 2025 | 7 Aug 2025 |
| Context | 200,000 | 400,000 |
| Input $/1M | $3.00 | - |
| Output $/1M | $15.00 | - |
| Cache read $/1M | - | $0.125 |
| Cache write $/1M | - | - |
| Speed | 0 tok/s | 0 tok/s |
| Latency | 0.00 s | 0.00 s |
| Modalities | text, image | text, image, file |
| Providers | - | openai |
| Publisher | Anthropic | OpenAI |
| Reported scores |
FAQ
- Is Claude 4 Sonnet or GPT-5 better?
- Across the 25 benchmarks both models report on Sophon, GPT-5 leads on 19 of them. Which one is "better" depends on the benchmark - the table above shows every shared score.
- How do Claude 4 Sonnet and GPT-5 compare on GPQA Diamond?
- Claude 4 Sonnet scores 68.3% and GPT-5 scores 68.6% on GPQA Diamond.
- When were Claude 4 Sonnet and GPT-5 released?
- Claude 4 Sonnet was released 22 May 2025 by Anthropic. GPT-5 was released 7 Aug 2025 by OpenAI.
Comparing something else? Build your own side-by-side.