Codex adds GPT‑5.1‑Codex family

OpenAI’s Codex documentation now lists support for a GPT‑5.1‑Codex model family, with a ‘Max’ option as the default and a ‘Mini’ option available, and it also notes GPT‑4o is not available within Codex. The change signals a segmentation of coding assistance into model families with different performance and cost trade‑offs. (help.openai.com)

OpenAI’s Codex docs now show a dedicated GPT‑5.1‑Codex family, with Max set as the default model inside the coding tool. (help.openai.com) The Help Center article says Codex “currently supports the GPT‑5.1‑Codex model family,” with Max as the default and Mini as an optional choice. The same page says retired or removed models, including GPT‑4o, “cannot be restored or purchased as a legacy tier.” (help.openai.com) Codex is OpenAI’s coding agent inside ChatGPT and in the command line interface, where it can inspect a repository, edit files, and run commands in a selected directory. OpenAI says Codex is included with ChatGPT Plus, Pro, Business, Edu, and Enterprise plans. (help.openai.com, help.openai.com) A model family is OpenAI’s way of grouping versions aimed at one job, in this case software work rather than general chat. The API docs describe GPT‑5.1‑Codex as “optimized for agentic coding tasks in Codex or similar environments.” (developers.openai.com) OpenAI has been separating coding models from general models across its product line for months. Its model release notes now tell ChatGPT users to use GPT‑5‑codex for coding-focused work and GPT‑5 for non-coding tasks, while March 11, 2026 release notes say GPT‑5.1 chat models were retired from ChatGPT. (help.openai.com) The pricing pages point to the same split between speed, cost, and capability. The GPT‑5.1‑Codex API page lists the standard model at $1.25 per million input tokens and $10 per million output tokens, while OpenAI’s Codex prompting guide says its starter prompt began as the default prompt for GPT‑5.1‑Codex‑Max. (developers.openai.com, developers.openai.com) OpenAI has also published a separate launch post for GPT‑5.1‑Codex‑Max, describing it as a model built for “long-running, project-scale work” with support for larger coding tasks and multi-hour agent loops. A separate system card says the model was reviewed under OpenAI’s Preparedness Framework and paired with product controls such as sandboxing and configurable network access. (openai.com, openai.com) The immediate effect for ChatGPT users is narrower than a brand-new product launch: Codex now presents coding help as a menu of specialized GPT‑5.1 variants rather than a mix of older general-purpose models. That makes Max the default lane for heavier coding work, Mini the lighter option, and GPT‑4o a model that no longer exists inside Codex. (help.openai.com)

Get your own daily briefing

Scout delivers personalized news, insights, and conversations tailored to your role and industry.

Download on the App Store

Shared from Scout - Be the smartest in the room.