Best fit
Implementation, review, and follow-up tasks where OpenAI coding models are the preferred agent.
Run Codex through an inspectable review path.
Codex handles implementation, review, and follow-up changes while ProPR keeps the task planned, isolated, logged, and attached to a pull request. It fits terminal-first work through the ProPR CLI.
Approved planContext and scope attached
GPT coding models
Isolated worker
GitHub pull requestCommits, logs, and follow-ups
Your credentials
Task history kept
GitHub merge decision
Agent fit and access
Access
Run on a subscription, or a direct API key.
Role in ProPR
Select Codex per phase or per issue. Task state stays in ProPR; iteration stays in GitHub.
Pick the OpenAI model that fits the phase.
ProPR's current Codex registry starts with the GPT-5.6 generation — Sol for the hardest work, Terra balanced, Luna for fast passes — with the GPT-5.5 and earlier registrations still available. A reasoning level can be set globally, per model, or per issue with a level-* label — from low to ultra, adapted to the Codex CLI. Your OpenAI account determines which registered models it can run through a ChatGPT subscription or API key.
| Model | OpenAI model ID | Context | Reach for it when |
|---|---|---|---|
| GPT-5.6 Sol / Terra / Luna | gpt-5.6-sol / -terra / -luna | 1M / 400K / 400K | The current generation: Sol for the hardest work, Terra balanced, Luna for fast, low-cost passes. |
| GPT-5.5 / GPT-5.5 Pro | gpt-5.5 / gpt-5.5-pro | 1.05M | Complex implementation, refactors, debugging, review, and frontend-heavy work. |
| GPT-5.4 / GPT-5.4 Pro | gpt-5.4 / gpt-5.4-pro | 1.05M | Strong general coding work with an established 5.4 configuration. |
| GPT-5.4 Mini / Nano | gpt-5.4-mini / gpt-5.4-nano | 400K | Fast, inexpensive passes and high-volume follow-ups. |
| GPT-5.3 Codex / Codex Spark | gpt-5.3-codex / gpt-5.3-codex-spark | 400K | Codex-tuned work, code review, or Spark research-preview testing. |
| GPT-5.2, GPT-5 Mini / Nano | gpt-5.2, gpt-5-mini / gpt-5-nano | 400K | Existing configurations that have not moved to the newer registered models. |
Older Codex-specific IDs such as gpt-5-codex, gpt-5.1-codex, gpt-5.1-codex-max, gpt-5.1-codex-mini, and gpt-5.2-codex are deprecated by OpenAI.
Bring your own OpenAI plan — ProPR never marks up tokens.
Codex runs on a ChatGPT subscription or a pay-as-you-go OpenAI API key. You pay OpenAI directly; ProPR adds no markup and resells nothing.
| Plan | Price | Good for |
|---|---|---|
| ChatGPT Plus | $20 / mo | Entry tier for light, single-developer use. |
| ChatGPT Pro 5× | $100 / mo | 5× Plus limits — sustained Codex work. |
| ChatGPT Pro 20× | $200 / mo | 20× Plus limits — heavy, multi-task days. |
| OpenAI API | Pay-as-you-go | Direct metered per-token billing when you prefer it. |
Prices as published by OpenAI, July 2026.
Model choice
Pick the OpenAI coding model that fits the phase.Phase routing
Route Codex at implementation and fast follow-up passes — the phases where its speed pays — and keep the PR review path unchanged.Access model
Run on a subscription, or a direct API key.Usage tracking
Enable Agent Tank and ProPR tracks Codex usage around task calls.
Setup details live in the docs.
ProPR runs the official Codex CLI (@openai/codex), and the setup specifics — CLI versions, model aliases, and credentials — are covered in the documentation. Inside ProPR, Codex follows the same path as every other agent: planning, isolated implementation, pull request, follow-up, and recovery. See the direct Codex comparison for the operating boundary.