GPT-5.1 Codex Mini (High) — benchmark results

Provider: OpenAI. Released 2025-11-19. Access: API.

Unified ELO 1729 ± 23, rank #226 of 1841 rated models, from 53 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
AA LiveCodeBench83.6Pass@1 (%)94.7
AA AIME 202591.67Accuracy (%)93.1
LLM Chess (Saplin)544ELO84.3
AA IFBench67.89Accuracy (%)81.4
Epoch AI - Scicode42.59Score78.9
AA MMLU-Pro82Accuracy (%)77.9
AA Terminal-Bench Hard33.33Accuracy (%)77.7
Artificial Analysis Intelligence Index30.63Intelligence Index75.8
AA Long Context Reasoning62.67Accuracy (%)75.2
AA Global-MMLU-Lite - Yoruba69.92Accuracy (%)74.8
AA GPQA Diamond81.31Accuracy (%)74.7
AA Humanity's Last Exam16.91Accuracy (%)74.7

Interactive version: theaggregate.ai/model?slug=gpt-5-1-codex-mini-high · How It Works · Data refreshed daily, snapshot 2026-07-25.