DeepSeek V3.2 vs GPT-5 Nano
Head to head on 22 shared benchmarks, with price, context window, and release dates.
Benchmark scores
60evals · sort by any column, ⤢ to expand fullscreen
| Model | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| DeepSeek V3.2 DeepSeek | 59.0% | 1511elo | 66.8 | 22.1% | 75.1% | 11.2% | 49.0% | 65.9% | 64.6% | 50.0% | 48.2% | 70.4% | 85.0% | 77.2% | 59.3% | 83.7% | 8.0% | 38.7% | 70.0% | 68.2% | 32.6% | 78.9% | 74.7% | - | 74.2% | - | - | - | 0.0% | 47.9% | 16.0% | - | - | - | - | 31.3% | 68.0% | - | - | - | - | 57.5% | - | - | - | - | - | - | - | - | - | 59.0% | 65 | - | - | - | 5.1% | - | - | - | 51.7% |
| GPT-5 Nano OpenAI | 27.3% | 703elo | 28 | 8.3% | 42.8% | 4.0% | 32.5% | 52.8% | 62.4% | 43.4% | 55.7% | 46.8% | 68.4% | 40.3% | 47.0% | 55.6% | 12.0% | 29.1% | 34.8% | 67.4% | 6.8% | 25.7% | - | 0.0% | - | 0.25 | 0.21 | 6.7% | - | - | - | 10.0% | 78.2% | 1.18 | 100.0% | - | - | 1.38 | 1.38 | 100.0% | 100.0% | - | 26.7% | 30.4% | 72.9% | 53.6% | 100.0% | 0.0% | 0.0% | 30.4% | 1.72 | - | - | 29.7% | 20.41 | -6.67 | - | 82.0% | 82.0% | 40.0% | 43.7% |
2 / 2 models
| Attribute | DeepSeek V3.2 DeepSeek | GPT-5 Nano OpenAI |
|---|---|---|
| Description | DeepSeek V3.2 is an AI model from DeepSeek, released with open weights. | GPT-5 nano is an AI model from OpenAI. |
| Family | Deepseek | GPT |
| Params | - | - |
| Open weights | open | closed |
| License | mit | proprietary |
| Released | 1 Dec 2025 | 7 Aug 2025 |
| Context | 163,840 | 400,000 |
| Input $/1M | $0.28 | $0.05 |
| Output $/1M | $0.42 | $0.40 |
| Cache read $/1M | $0.130 | $0.005 |
| Cache write $/1M | - | - |
| Speed | 0 tok/s | 0 tok/s |
| Latency | 0.00 s | 0.00 s |
| Modalities | text | text, image, file |
| Providers | - | - |
| Publisher | DeepSeek | OpenAI |
| Reported scores | Aider Polyglot Benchmark74.2% (Aider) MathArena57.47% GPQA Diamond75.1% (AA) Humanity's Last Exam (HLE)11.2% (AA) |
FAQ
- Is DeepSeek V3.2 or GPT-5 Nano better?
- Across the 22 benchmarks both models report on Sophon, DeepSeek V3.2 leads on 20 of them. Which one is "better" depends on the benchmark - the table above shows every shared score.
- How do DeepSeek V3.2 and GPT-5 Nano compare on GPQA Diamond?
- DeepSeek V3.2 scores 75.1% and GPT-5 Nano scores 42.8% on GPQA Diamond.
- Which is cheaper, DeepSeek V3.2 or GPT-5 Nano?
- GPT-5 Nano is cheaper at $0.40 per million output tokens against $0.42 for DeepSeek V3.2 - about 1.0x.
- When were DeepSeek V3.2 and GPT-5 Nano released?
- DeepSeek V3.2 was released 1 Dec 2025 by DeepSeek. GPT-5 Nano was released 7 Aug 2025 by OpenAI.
Comparing something else? Build your own side-by-side.