IFBench ↗77.62%
GPT-5.2 Codex (Xhigh) · IFBench · AA implementation ↗
OpenAI / MODEL EVIDENCE
GPT-5.2 Codex
7 benchmarks · 7 observations · 7 sources
Scores at a glance
7 benchmarks · large dot: GPT-5.2 Codex; small dots: closest models where reported
Terminal-Bench 2.0 ↗66.52%
GPT-5.2 Codex · Terminal Bench · Accuracy mean · % ↗
Text Arena (Coding) ↗1338.7 Elo
GPT-5.2 Codex · WebDev Arena · Elo ↗
Vibe Code Bench ↗37.91 %
GPT 5.2 Codex (OpenHands) · Vibe Code Bench · original task set ↗
SWE-bench Verified ↗72.4 %
GPT 5.2 Codex (Mini-SWE-agent) · SWE-bench Verified · Score ↗
LiveCodeBench ↗87.99 %
GPT 5.2 Codex · LiveCodeBench · Accuracy ↗
NYU CTF Bench ↗61.7%
GPT-5.2 Codex · NYU CTF Bench · Cotool 81-challenge subset · accuracy ↗
Closest models by score
Similar positions on shared benchmark leaderboards. Smaller gaps are closer.
There are not enough shared scored settings to identify close comparisons reliably.
Sources
Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.
- Arena leaderboard — WebDev evaluation ↗
- IFBench Benchmark | Artificial Analysis ↗
- NYU CTF Bench — Cotool evaluation, February 2026 ↗
- Terminal-Bench 2.0 leaderboard ↗
- LiveCodeBench — evaluation methodology and results ↗
- SWE-bench Verified — evaluation methodology and results ↗
- Vibe Code Bench — evaluation methodology and results ↗
Release dates reported by sources
2025-12-11 · 2025-12-18
Names used in source tables
GPT 5.2 Codex · GPT 5.2 Codex (Mini-SWE-agent) · GPT 5.2 Codex (OpenHands) · GPT-5.2 Codex · GPT-5.2 Codex (Xhigh)