Up to four models
GPT-5.2 vs DeepSeek-V3.2
On the index (GPT-4 = 100), GPT-5.2 scores 162 and DeepSeek-V3.2 141. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
GPT-5.2
162146–178
WeightsClosed
ReleasedDec 2025
Price in / out$1.75 / $14
Context400K
Core tests taken6 of 9
141125–157
WeightsOpen
ReleasedDec 2025
Price in / out$0.26 / $0.38
Context164K
Core tests taken2 of 9
GPT-5.2 leads DeepSeek-V3.2 by 21 points, ahead on 1 of 1 shared core tests, at 17× the price.
Test by testthird-party results
GPT-5.2
DeepSeek-V3.2
Every result as a table
| Test | GPT-5.2 | DeepSeek-V3.2 |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 91.4% | 83.4% |
| CL-bench | 18.2% | 12.4% |
| Maths | ||
| AIME-style maths (OTIS mock) | 96.1% | 87.8% |
| FrontierMath (2025 set) | 40.7% | 22.1% |
| FrontierMath tier 4 (2025 set) | 18.8% | 2.1% |
| Coding | ||
| Terminal-Bench 2.0 (Terminus 2 runs) | 54.0% | 39.6% |
| Agents | ||
| DTBench | 90.9% | 85.6% |
| LMCA | 43.9% | 28.8% |
Task length · METR45 sec
Tests at least two of these models have taken. Bold rows are the core tests.