name = "GPT-5.3 Codex"
description = "Coding-optimized GPT model for repository edits, reviews, and agentic software work"
family = "gpt-codex"
release_date = "2026-02-05"
last_updated = "2026-02-05"
attachment = true
reasoning = true
temperature = false
tool_call = true
structured_output = true
knowledge = "2025-08-31"
open_weights = false

[limit]
context = 400_000
input = 272_000
output = 128_000

[modalities]
input = ["text", "image", "pdf"]
output = ["text"]

[[benchmarks]]
name = "SWE-Atlas Codebase QnA"
score = 32.6
metric = "score"
harness = "Codex"
source = "https://labs.scale.com/leaderboard/sweatlas-qna"

[[benchmarks]]
name = "SWE-Atlas Refactoring"
score = 42.38
metric = "score"
harness = "Codex"
source = "https://labs.scale.com/leaderboard/sweatlas-refactoring"

[[benchmarks]]
name = "SWE-Atlas Test Writing"
score = 38.98
metric = "score"
harness = "Codex"
source = "https://labs.scale.com/leaderboard/sweatlas-tw"