GPT-5.2 Instant: benchmark results
Provider: OpenAI. Released 2025-12-11. Access: API.
Unified ELO 1740 ± 18, rank #162 of 1605 rated models, from 82 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| PhysAI-Bench | 49.4 | Decision accuracy (%; four-option choice of the next action | 96.4 |
| Chatbot Arena (Text - Multi-Turn) | 1494 | Arena Score | 96 |
| Chatbot Arena (Text - Medicine & Healthcare) | 1500 | Arena Score | 95.9 |
| Chatbot Arena (Text - Business, Management & Financial Ops) | 1484 | Arena Score | 95.1 |
| Chatbot Arena (Text - Spanish) | 1476 | Arena Score | 94.9 |
| LM Market Cap LMC Score | 90.5 | LMC Score (0-100) | 94.7 |
| UGI - Natural Intelligence | 61.46 | NatInt Score | 94.3 |
| Chatbot Arena (Text) | 1476 | Elo | 93.5 |
| Chatbot Arena (Text - Hard Prompts (English)) | 1499 | Arena Score | 93.1 |
| Chatbot Arena (Text - Software & IT Services) | 1510 | Arena Score | 93 |
| UGI - Writing | 60.78 | Writing Score | 93 |
| Chatbot Arena (Text - English) | 1481 | Arena Score | 92.9 |
Interactive version: theaggregate.ai/model?slug=gpt-5-2-instant · How It Works · Data refreshed daily, snapshot 2026-09-26.