GPT-5.2

OpenAI·gpt-5-2

Rank · overall#119

Score57.4

Benchmark scores

11 benchmarks

SWE-bench Verified72.8%

% Resolved · 2025-12-11

Chatbot Arena (Coding)1470Coding Elo

ARC-AGI52.9%

Score

WebDev Arena1418WebDev Elo · ± 23

FrontierMath (Tier 4)31.7%

Accuracy · ± 7.3

SWE-bench Pro29.9%

% Resolved

SWE-bench Multilingual66.7%

% Resolved · 2026-02-13

MCP Atlas67.6Score

SkillsBench51.7Score

MathArena Apex13.5%

Accuracy

BrowseComp65.8%

Accuracy

0 agents

This model appears only in model-level benchmarks — no harness ran it directly.

Scraped and aggregated from public leaderboards · 2026-08-16