GPT-5.2
#42 of 274 scored.
OpenAIDec 11, 2025Closed
Capabilities Index
Range 151.6–155.4
153.4
Context
Takes file, image, text
400K
Input
Per 1M tokens
$1.75
Output
Per 1M tokens
$14
Benchmarks
How it scores.
- GPQA Diamond88.5%
27th of 186
- SWE-bench Verified73.8%
17th of 32
- FrontierMath67.4%
26th of 81
- ARC-AGI-252.9%
30th of 79
- Humanity's Last Exam24.2%
12th of 40
- SimpleQA Verified37.1%
47th of 77
- Mock AIME96.1%
29th of 176
- Terminal-Bench64.9%
8th of 35
Everything else Epoch has for it
- ARC-AGI86.2%
- DTBench84.9%
- VPCT76.0%
- METR Time Horizons75.3%
- WeirdML72.2%
- LMCA51.7%
- GDPval49.7%
- Chess Puzzles46.3%
- DeepResearch Bench41.1%
- SimpleBench35.0%
- FrontierMath-Tier-4-v2-Private31.7%
- GSO-Bench27.5%
- EBR-bench23.0%
- CL-bench18.2%
- Mystery Game Puzzles15.2%
- ProofBench15.0%
- Furniture Assembly11.9%
- Remote Labor Index2.5%
On the index
Around it.
Sources