OpenAI·released 2025-12-111 source
GPT-5.2 Pro benchmark scores: 6 benchmarks tracked. 33% of its scores are independently reproduced — the rest are self-reported or unverified.
Leads on 0 of 6 benchmarks·best result 90.5 on ARC-AGI (unverified)·2 of 6 independently reproduced·$21/$168 per M tokens
Consensus: LiteLLM · Cross-check: models.dev, OpenRouter · See every model’s pricing →
| Benchmark | Score | Evidence status | Measured |
|---|---|---|---|
| ARC-AGI | 90.5accuracy (%) | unverified· optimizedT1 | 2025-12-11https://arcprize.org/leaderboard |
| FrontierMath-Tiers-1-3-v2-Private | 74accuracy (%) | reproducedT1 | 2025-12-11Epoch AI |
| Chess Puzzles | 57.91accuracy (%) | unverifiedT2 | 2025-12-11 |
| ARC-AGI-2 | 54.16accuracy (%) | unverifiedT2 | 2025-12-11 |
| SimpleBench | 48.88accuracy (%) | unverifiedT1 | 2025-12-11https://lmcouncil.ai/benchmarks |
| FrontierMath-Tier-4-v2-Private | 46accuracy (%) | reproducedT1 | 2025-12-11Epoch AI |
Claims drawn from cited facts, not live model generation.
This model was developed by OpenAI and released in 2025.
2 cited facts
This model has scores on 6 tracked benchmarks and currently leads 0 of them. The record’s evidence status is independently reproduced.
3 cited facts
Its mean gap to the best recorded score across its tracked benchmarks is 27.78 points.
1 cited fact
5 facts cross-checked across data sources: 2 corroborated, 1 single-source, 2 disputed.
Source facts, citations, and refresh stamp for this record.
Sources: epoch_benchmarks, litellm_prices, modelsdev_models, openrouter_models
GPT-5.2 Pro is an AI model developed by OpenAI, released 2025-12-11. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.
GPT-5.2 Pro has recorded scores on 6 benchmarks, each shown with its evidence status.
GPT-5.2 Pro has recorded scores on 6 benchmarks — Chess Puzzles, SimpleBench, ARC-AGI, ARC-AGI-2, FrontierMath-Tiers-1-3-v2-Private, FrontierMath-Tier-4-v2-Private. The full table above shows each score with its evidence status.
2 of 6 recorded scores (33%) are independently reproduced rather than self-reported by the lab.
GPT-5.2 Pro has 6 tracked claims: 2 independently reproduced, 4 unverified.
Listed API pricing: $21 per million input tokens, $168 per million output tokens (prices disputed across sources). See the pricing block for the full breakdown.