GPT-5.1 Chat: benchmark results

Provider: OpenAI. Released 2025-11-12. Access: API.

Unified ELO 1717 ± 31, rank #295 of 2656 rated models, from 14 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
BenchTable - STEM77.6Weighted Score (%)89.2
BenchTable - Reasoning67.9Weighted Score (%)83.4
BenchTable65.6Total Score (%)79.4
Korean CSAT 2026 (Easy Mode) - English97Points (out of 100)66
BenchTable - Utility60.1Weighted Score (%)62.8
BenchTable - Tech61.7Weighted Score (%)59.4
SnakeBench23.3TrueSkill Rating54.7
Korean CSAT 2026 (Easy Mode) - Total372.5Points (out of 450)33.7
Korean CSAT 2026 (Easy Mode) - Physics I23Points (out of 50)31.6
Korean CSAT 2026 (Easy Mode) - Mathematics88Points (out of 100)26.9
Korean CSAT 2026 (Easy Mode) - Society and Culture30Points (out of 50)24.8
Korean CSAT 2026 (Easy Mode) - Life Science I30Points (out of 50)23.8

Interactive version: theaggregate.ai/model?slug=gpt-5-1-chat · How It Works · Data refreshed daily, snapshot 2026-09-19.