DeepSeek V3.2 vs GPT-4o
Head to head on 18 shared benchmarks, with price, context window, and release dates.
Benchmark scores
48evals · sort by any column, ⤢ to expand fullscreen
| Model | |||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| DeepSeek V3.2 DeepSeek | 74.2% | 59.0% | 47.9% | 22.1% | 75.1% | 11.2% | 49.0% | 65.9% | 64.6% | 48.2% | 70.4% | 59.3% | 83.7% | 38.7% | 70.0% | 68.2% | 32.6% | 78.9% | 74.7% | - | - | - | - | 0.0% | 16.0% | - | 1511elo | 66.8 | - | 31.3% | 68.0% | - | 50.0% | 85.0% | 77.2% | - | 57.5% | - | - | - | 8.0% | - | 59.0% | - | 65 | - | 5.1% | - | 51.7% |
| GPT-4o OpenAI | 18.2% | 6.0% | 45.9% | 0.3% | 54.3% | 2.4% | 34.3% | 55.1% | 46.1% | 70.1% | 49.0% | 30.9% | 74.8% | 33.3% | 21.6% | 74.5% | 8.3% | 25.1% | - | 15.0% | 38 | 0.0% | 0.33 | - | - | 6.1% | - | - | 96.1 | - | - | 0.0% | - | - | - | 75.9% | - | 10.8% | 57.4% | 54.0% | - | 5.71 | - | 61.2 | - | 0.98 | - | 32.5% | 33.4% |
2 / 2 models
| Attribute | DeepSeek V3.2 DeepSeek | GPT-4o OpenAI |
|---|---|---|
| Description | DeepSeek V3.2 is an AI model from DeepSeek, released with open weights. | GPT-4o is an AI model from OpenAI. |
| Family | Deepseek | GPT |
| Params | - | - |
| Open weights | open | closed |
| License | mit | proprietary |
| Released | 1 Dec 2025 | 20 Nov 2024 |
| Context | 163,840 | 128,000 |
| Input $/1M | $0.28 | $2.50 |
| Output $/1M | $0.42 | $10.00 |
| Cache read $/1M | $0.130 | $1.250 |
| 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) | Taubench61.2 pass^1 GSM8K96.1 Accuracy ScholarSearch5.71 Social Sciences & Humanities (%) MortgageTax57.4% AIME202438 pass@1 |
FAQ
- Is DeepSeek V3.2 or GPT-4o better?
- Across the 18 benchmarks both models report on Sophon, DeepSeek V3.2 leads on 16 of them. Which one is "better" depends on the benchmark - the table above shows every shared score.
- How do DeepSeek V3.2 and GPT-4o compare on GPQA Diamond?
- DeepSeek V3.2 scores 75.1% and GPT-4o scores 54.3% on GPQA Diamond.
- Which is cheaper, DeepSeek V3.2 or GPT-4o?
- DeepSeek V3.2 is cheaper at $0.42 per million output tokens against $10.00 for GPT-4o - about 23.8x.
- When were DeepSeek V3.2 and GPT-4o released?
- DeepSeek V3.2 was released 1 Dec 2025 by DeepSeek. GPT-4o was released 20 Nov 2024 by OpenAI.
Comparing something else? Build your own side-by-side.