Skip to main content
AI Choice Engine

OpenAI

GPT-5.3-Codex

GPT-5.3-Codex is a OpenAI model with 400K context, $1.75 in / $14 out per million tokens, last verified October 25, 2026. Open weights: no. Status: Superseded since October 1, 2026; the current pick is GPT-6 Sol. Public-board scores are linked to the publisher; vendor-only figures stay in the watch-out, not on the board.

OpenAI’s GPT-5.3 coding-specialized API model for Codex or similar agentic environments, with a 400K-token context at $1.75/$14.

Recorded profile verification October 25, 2026Individual provider fields may changeAccess: deprecated — Deprecated October 1, 2026; scheduled API shutdown April 1, 2027. OpenAI recommends gpt-6-sol, a separately served model.

AI model specification details
SpecificationGPT-5.3-Codex
ProviderOpenAI
TierFrontier
Context window400K
Max output128K
Input / 1M tokens$1.75
Output / 1M tokens$14
WeightsClosed
ParametersNot verifiedUnverified
Reasoning levelslow, medium, high, xhigh
Modalitiestext, image
API model idgpt-5.3-codex
ReleasedFebruary 5, 2026

Pricing tiers: All amounts are USD per MTok. Standard: input/output $1.75/$14.00, cached input $0.175. Fast: input/output $3.50/$28.00, cached input $0.35. Regional processing and FedRAMP add 10% where the model and route are eligible; region eligibility is separate from a quoted tariff. Server-side tools and their tokens can add charges. The specialized-model table now quotes Fast $3.50/$28 with $0.35 cached input; no Batch/Flex or write rate for this exact ID is established here. Deprecated October 1, 2026; scheduled API shutdown April 1, 2027, replacement gpt-6-sol.

Best for

  • Autonomous coding
  • Repo-scale refactors
  • Test generation

Watch out

Deprecated October 1, 2026; migrate to gpt-6-sol before the announced April 1, 2027 API shutdown. Reasoning supports low/medium/high/xhigh; none and max are not listed. A model ID is distinct from the Codex application and its current selectable models; no automatic routing to a successor is established.

History

GPT-5.3-Codex timeline

Release and status events for this model that passed two-source verification — dated, sourced, and linked.

  • GPT-5.3-Codex API deprecationStatus updateOctober 1, 2026

    OpenAI deprecates gpt-5.3-codex, scheduling API removal for April 1, 2027.

    OpenAI API deprecations

Source receipts

Catalog figures for this model were checked against the following sources.

Common questions

GPT-5.3-Codex

Answered from the verified figures on this page rather than general guidance.

How much does GPT-5.3-Codex cost per million tokens?
GPT-5.3-Codex lists base rates of $1.75 per million input tokens and $14 per million output tokens. Estimate both input and output token volume and any additional charges for your workload. All amounts are USD per MTok. Standard: input/output $1.75/$14.00, cached input $0.175. Fast: input/output $3.50/$28.00, cached input $0.35. Regional processing and FedRAMP add 10% where the model and route are eligible; region eligibility is separate from a quoted tariff. Server-side tools and their tokens can add charges. The specialized-model table now quotes Fast $3.50/$28 with $0.35 cached input; no Batch/Flex or write rate for this exact ID is established here. Deprecated October 1, 2026; scheduled API shutdown April 1, 2027, replacement gpt-6-sol.
What is GPT-5.3-Codex's context window?
GPT-5.3-Codex accepts about 400K tokens of context. That only matters if you routinely send very long documents, large codebases, or multi-turn histories that approach that limit.
What is GPT-5.3-Codex best for?
GPT-5.3-Codex is a frontier tier from OpenAI. It suits autonomous coding, repo-scale refactors, test generation. Deprecated October 1, 2026; migrate to gpt-6-sol before the announced April 1, 2027 API shutdown. Reasoning supports low/medium/high/xhigh; none and max are not listed. A model ID is distinct from the Codex application and its current selectable models; no automatic routing to a successor is established.
Model comparison catalogBenchmark library