Skip to main content

Access OpenAI models via a Codex subscription

Project description

llm-openai-codex

LLM plugin for accessing ChatGPT/Codex-backed OpenAI models through the Responses API.

This project is forked from and based on Simon Willison's llm-openai-via-codex.

Changes from the original plugin

  • Package renamed to llm-openai-codex.
  • Model prefix changed to codex/.
  • Plugin-owned auth is stored in LLM's user config directory as auth-codex.json, with borrowed fallback to ${CODEX_HOME:-~/.codex}/auth.json.
  • llm codex auth commands manage login, import, status, refresh, and logout.
  • Explicit verbosity option maps to Responses API text.verbosity.
  • Extra Responses API options are forwarded when LLM accepts them.
  • Missing account_id values are derived from OAuth JWT claims when possible.
  • Added llm codex usage showing current Codex plan usage.
  • Registers known hidden/fallback Codex models such as gpt-5.3-codex-spark.

Installation

Install this plugin in the same environment as LLM.

llm install llm-openai-codex

Usage

Authenticate the plugin:

llm codex login [--device-code]

If you already use Codex CLI, the plugin can fall back to ${CODEX_HOME:-~/.codex}/auth.json when auth-codex.json does not exist. You can also copy those tokens into plugin-owned auth with llm codex import.

List available Codex-backed models:

llm models -q codex

Models are discovered from the Codex API, plus known hidden/fallback slugs such as gpt-5.3-codex-spark. Some listed fallback models may still fail at request time if your plan lacks access.

Run a prompt:

llm -m codex/gpt-5.3-codex-spark "Hello"

Use Responses API verbosity:

llm -m codex/gpt-5.3-codex-spark -o verbosity low "Summarize this"

Auth commands

llm codex login
llm codex login --device-code
llm codex status
llm codex refresh
llm codex refresh --borrowed
llm codex usage
llm codex import
llm codex logout

Auth source order:

  1. Plugin-owned auth-codex.json
  2. Borrowed Codex CLI auth at ${CODEX_HOME:-~/.codex}/auth.json

llm codex import copies ChatGPT OAuth tokens from Codex CLI auth into plugin-owned auth-codex.json. It refuses to overwrite an existing auth-codex.json; run llm codex logout first if you want to replace plugin-owned auth.

Refresh behavior

  • Plugin-owned auth refreshes lazily and on demand via llm codex refresh.
  • Borrowed Codex CLI auth never refreshes automatically. When it expires, pick one:
    • llm codex refresh --borrowed — refresh the shared file in place. Rotates the shared refresh_token; restart any running Codex CLI session afterwards.
    • Run Codex CLI itself to refresh.
    • llm codex import to promote it to plugin-owned auth.

llm codex status shows which auth source is active.

When authentication is missing or expired, run llm codex login [--device-code] or llm codex import.

Development

uv run pytest
uv run llm plugins
uv run llm codex status

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

llm_openai_codex-0.2.3.tar.gz (22.5 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

llm_openai_codex-0.2.3-py3-none-any.whl (17.8 kB view details)

Uploaded Python 3

File details

Details for the file llm_openai_codex-0.2.3.tar.gz.

File metadata

  • Download URL: llm_openai_codex-0.2.3.tar.gz
  • Upload date:
  • Size: 22.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for llm_openai_codex-0.2.3.tar.gz
Algorithm Hash digest
SHA256 153288c4365be26d02ac0a708992e4d518a84b39a28ed3597262dcb6471572d8
MD5 4f93e2b1345c5f6c09024d40b240d8e9
BLAKE2b-256 2f4817223961597b6d02c3fafb594ea5a0cf17b21c91e1109dd159b69d3ddd56

See more details on using hashes here.

Provenance

The following attestation bundles were made for llm_openai_codex-0.2.3.tar.gz:

Publisher: publish.yml on rdslw/llm-openai-codex

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file llm_openai_codex-0.2.3-py3-none-any.whl.

File metadata

File hashes

Hashes for llm_openai_codex-0.2.3-py3-none-any.whl
Algorithm Hash digest
SHA256 b7c3bcb868f05bcbdfa3d6f90e0bfc4320e3c555529b857471dc4d405c78baff
MD5 3adec15c36a058f9dd40c066b4862445
BLAKE2b-256 34e10c2f6aa403d35b942afcc1976026448b01c542af3d7a7e335b640223249e

See more details on using hashes here.

Provenance

The following attestation bundles were made for llm_openai_codex-0.2.3-py3-none-any.whl:

Publisher: publish.yml on rdslw/llm-openai-codex

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page