GPT-5 Chat: benchmark results
OpenAI's non-reasoning, ChatGPT-tuned GPT-5 snapshot exposed in the API as gpt-5-chat. Provider: OpenAI. Released 2025-08-07. Access: API.
Unified ELO 1636 ± 1, rank #158 of 1392 rated models, from 88 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AmongAIs | 750 | Score (50×Impostor + 10×Crewmate wins) | 100 |
| AI Chess Leaderboard (Continuation) | 1465 | Elo | 95.1 |
| UGI - Writing | 62.69 | Writing Score | 94 |
| LLM Arena RU | 1129 | Arena Elo | 93.2 |
| UGI - Natural Intelligence | 57.12 | NatInt Score | 92.7 |
| BenchTable | 71.4 | Total Score (%) | 88.1 |
| SLR-Bench - Basic | 100 | Accuracy (%) | 86.4 |
| MMTU - KB mapping | 54.27 | Accuracy (%) | 84 |
| MMTU - NL-2-code | 70.42 | Accuracy (%) | 84 |
| Chatbot Arena (Text - Japanese) | 1432 | Arena Score | 82.7 |
| RAI-Bench - RAG Robustness (LC Abstention) | 89 | Rate (%) | 82.6 |
| SLR-Bench - Hard | 14 | Accuracy (%) | 80.9 |
Interactive version: theaggregate.ai/model?slug=gpt-5-chat · How It Works · Data refreshed daily, snapshot 2026-09-05.