gpt-5.3-codex

Vendor openai Tier standard Price per million tokens, in / out $1.75 / $14 Released 2026-02-05 Context 400,000 Max output 128,000 Vendor page

Last updated 2026-10-26 How this is made

Meant to excel at

No sourced claims yet.

Watch out for

No sourced claims yet.

Benchmarks

Agentic coding percentile 0.33

Benchmark	Score	Percentile	Source	Date
swe-bench pro	56.8	0.33	Vendor's own site openai.com	2026-02-05

Computer use percentile 0.00

Benchmark	Score	Percentile	Source	Date
osworld-verified	64.7	0.00	Vendor's own site openai.com	2026-02-05

Percentile is the share of tracked models this model beats on that benchmark, averaged across the group. Only benchmarks at least four tracked models share are scored.

Practitioner reports

+/-

Poster complained that after earlier removals they had moved from Codex 5.3 to 5.5 extended, indicating 5.3-Codex had been withdrawn from their Codex setup; no performance claims about 5.3-Codex itself.

Task: agentic-long-running | Compared with gpt-5.5 | 2026-07-12 older than 90 days
Practitioner report community.openai.com