feat: add chatgpt 5.3-codex provider model (chatgpt/gpt-5.3-codex)

[vincentkoc]

Relevant issues

Adding OpenAI 5.3-codex (ChatGPT provider only) with baseline token limits since OpenAI has not released model on API yet. Another PR is drafted and will be updated once API is released #20550

Pre-Submission checklist

Please complete all items before asking a LiteLLM maintainer to review your PR

• I have Added testing in the tests/litellm/ directory, Adding at least 1 test is a hard requirement - see details
• My PR passes all unit tests on make test-unit
• My PR's scope is as isolated as possible, it only solves 1 specific problem

CI (LiteLLM team)

CI status guideline:

• 50-55 passing tests: main is stable with minor issues.
• 45-49 passing tests: acceptable but needs attention
• <= 40 passing tests: unstable; be careful with your merges and assess the risk.

• Branch creation CI run
  Link:

• CI run for the last commit
  Link:

• Merge / cherry-pick CI run
  Links:

Type

🆕 New Feature

Changes

Added new chatgpt/gpt-5.3-codex model entry

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
litellm	Ready	Preview, Comment	Feb 6, 2026 5:05am

[greptile-apps]

Greptile Overview

Greptile Summary

Added configuration for chatgpt/gpt-5.3-codex model to support OpenAI's upcoming GPT-5.3 Codex variant available through ChatGPT subscription

Changes

• Added chatgpt/gpt-5.3-codex entry to model_prices_and_context_window.json
• Configuration mirrors chatgpt/gpt-5.2-codex spec: 128k input/output tokens, responses mode, function calling support
• Previous JSON syntax error (missing comma) has been corrected

Notes

• PR description claims tests were added, but no test files were modified in this PR
• Model not yet available in OpenAI API (as noted in PR description)

Confidence Score: 5/5

• Safe to merge - simple configuration addition with correct JSON syntax
• Configuration correctly mirrors existing gpt-5.2-codex spec, JSON syntax is valid, and change is isolated to model registry
• No files require special attention

Important Files Changed

Filename	Overview
model_prices_and_context_window.json	Added chatgpt/gpt-5.3-codex model configuration with 128k token limits, matching 5.2-codex spec

Sequence Diagram

sequenceDiagram
    participant User
    participant LiteLLM
    participant ModelConfig
    participant ChatGPT

    User->>LiteLLM: Request with model="chatgpt/gpt-5.3-codex"
    LiteLLM->>ModelConfig: Load model_prices_and_context_window.json
    ModelConfig-->>LiteLLM: Return config (128k tokens, /v1/responses endpoint)
    LiteLLM->>LiteLLM: Validate token limits (max_input: 128k, max_output: 128k)
    LiteLLM->>ChatGPT: Forward request to /v1/responses endpoint
    ChatGPT-->>LiteLLM: Stream response
    LiteLLM-->>User: Return formatted response

Loading

[greptile-apps]

_{1 file reviewed, 1 comment}

_{Edit Code Review Agent Settings | Greptile}

[greptile-apps]

_{1 file reviewed, 1 comment}

_{Edit Code Review Agent Settings | Greptile}

[greptile-apps]

_{1 file reviewed, no comments}

_{Edit Code Review Agent Settings | Greptile}

[hedeqiang]

Please merge this PR, thank you very much.

[vincentkoc]

Superseded by #22734, which refreshes ChatGPT OAuth 5.3 aliases on top of latest main and consolidates the outdated #20550/#20552 work. Closing stale PR to avoid split review.