Benchfolio

Leaderboard / Model

GPT-5.3 Codex

OpenAI · Closed weights · Released Feb 5, 2026 · $1.75 in · $14 out per 1M tokens

n/aOverall · not enough data

Score by job

Coding

Not enough data · 2 of 5 benchmarks

BenchmarkResultVariantz
Terminal-Bench78.4%gpt-5.3-codex0.75
SWE-bench Verified74.8%gpt-5.3-codex_high-0.47

Agents

Not enough data · 1 of 4 benchmarks

BenchmarkResultVariantz
METR Time Horizons5.8 hgpt-5.3-codex-0.08

Reasoning

Not enough data · 0 of 4 benchmarks

Math

Not enough data · 0 of 5 benchmarks

Writing

Not enough data · 0 of 2 benchmarks

Design

8 · rank 43 · 1 of 1 benchmarks

BenchmarkResultVariantz
LMArena WebDev1409 ratinggpt-5.3-codex (codex-harness)-1.48