O1 Pro — benchmark results

OpenAI's pro-tier o1 variant using more inference compute. Provider: OpenAI. Released 2024-12-05. Access: API.

Unified ELO 1608 ± 35, rank #463 of 1776 rated models, from 24 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
OpenAI ChatGPT Pro - AIME 202486pass@1 accuracy (%)100
OpenAI ChatGPT Pro - AIME 2024 4/4 Reliability804/4 reliability (%)100
OpenAI ChatGPT Pro - Codeforces90pass@1 percentile100
OpenAI ChatGPT Pro - Codeforces 4/4 Reliability754/4 reliability percentile100
OpenAI ChatGPT Pro - GPQA Diamond79pass@1 accuracy (%)100
OpenAI ChatGPT Pro - GPQA Diamond 4/4 Reliability744/4 reliability (%)100
SEAL - Agentic Tool Use (Enterprise)67.01Score93.9
SEAL - Agentic Tool Use (Chat)61.42Score90.9
TrackingAI IQ Test80.39IQ Test Score (%)73.3
Visual-Language Understanding47.32Score (self-reported)71.7
SEAL - VISTA47.32Score71
LLM Stats (AIME 2024)86Score (%)69.8

Interactive version: theaggregate.ai/model?slug=o1-pro · How the rankings work · Data refreshed daily, snapshot 2026-07-22.