GPT-5.2 Codex: benchmark results

OpenAI's coding-specialized GPT-5.2 Codex model for software tasks. Provider: OpenAI. Released 2026-01-14. Access: API.

Unified ELO 1689 ± 1, rank #71 of 1392 rated models, from 60 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
LiveBench Code Completion86.96Score99.1
LM Market Cap LMC Score90.5LMC Score (0-100)94.5
Vals AI LiveCodeBench87.99Accuracy (%)94.4
SnakeBench32.4TrueSkill Rating91.6
BenchmarkList ECI140.73Capability Index (ECI)89.9
AISI Cyber CTF72Success Rate (%)89.3
AI Chess Leaderboard (Reasoning)1303Elo88.2
AI Chess Leaderboard (Continuation)1151Elo86
Terminal-Bench 2.066.5Accuracy (%)85.7
SWE-bench Verified72.8Resolved (%)84.8
LiveBench Table Reformat100Score83
LiveBench Consecutive Events89.02Score77.4

Interactive version: theaggregate.ai/model?slug=gpt-5-2-codex · How It Works · Data refreshed daily, snapshot 2026-09-05.