Up to four models
GPT-5.2 vs GPT-5.1
On the index (GPT-4 = 100), GPT-5.2 scores 162 and GPT-5.1 150. 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
GPT-5.1
150134–166
WeightsClosed
ReleasedNov 2025
Price in / out$1.25 / $10
Context400K
Core tests taken4 of 9
GPT-5.2 leads GPT-5.1 by 12 points, ahead on 4 of 4 shared core tests, at 1.4× the price.
Test by testthird-party results
GPT-5.2GPT-5.1
Every result as a table
| Test | GPT-5.2 | GPT-5.1 |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 91.4% | 87.6% |
| Humanity's Last Exam | 27.8% | 23.7% |
| GPQA Diamond (Vals AI) | 91.7% | 86.6% |
| MMLU Pro | 86.2% | 86.4% |
| MMMU Pro | 86.7% | 83.2% |
| CL-bench | 18.2% | 23.7% |
| SimpleBench | 45.8% | 53.2% |
| SimpleQA Verified | 37.1% | 48.0% |
| VPCT | 84.0% | 58.7% |
| Maths | ||
| AIME-style maths (OTIS mock) | 96.1% | 88.6% |
| MGSM | 94.0% | 93.0% |
| Chess puzzles | 48.0% | 42.0% |
| FrontierMath (2025 set) | 40.7% | 31.0% |
| FrontierMath tier 4 (2025 set) | 18.8% | 12.5% |
| Coding | ||
| SWE-bench Verified | 73.8% | 68.0% |
| Terminal-Bench 2.0 (Terminus 2 runs) | 54.0% | 47.6% |
| LiveCodeBench | 85.4% | 86.5% |
| SWE-bench | 75.8% | 69.8% |
| Vibe Code Bench v1.1 | 53.5% | 24.6% |
| GSO-Bench | 27.5% | 13.7% |
| WeirdML | 72.2% | 60.8% |
| Agents | ||
| DTBench | 90.9% | 90.1% |
| DeepResearch Bench | 41.1% | 42.8% |
| LMCA | 43.9% | 43.9% |
| Mystery game puzzles | 23.0% | 19.0% |
| Professional work | ||
| CorpFin v2 | 65.9% | 63.8% |
| LegalBench | 82.8% | 85.7% |
| MedCode | 49.8% | 52.7% |
| MedScribe | 84.4% | 88.1% |
| MortgageTax | 67.1% | 61.4% |
| TaxEval v2 | 75.8% | 74.9% |
| Novel problems | ||
| ARC-AGI-2 | 72.9% | 17.6% |
| ARC-AGI-1 | 94.5% | 72.8% |
Task length · METR45 sec
Tests at least two of these models have taken. Bold rows are the core tests.