Codex (VS Code extension) stops working mid-session with "model at capacity" errors

timore: > Would appreciate any additional insight This is pure speculation on my part but also based on intuition from past experience of using LLMs and being on this forum daily for a few years. A significant change in the last 24 hours was the release of gpt-image-2 Introducing gpt-image-2 - available today in the API and Codex Announcements > gpt-image-2 is OpenAI’s most capable image generation model yet. It is designed for complex visual tasks and produces precise, usable images with stronger editing, better layouts, improved text rendering, and more reliable instruction-following. [YouTube] Starting today, developers can use gpt-image-2 in the API and in Codex. This release is built for production workflows, where images need to be accurate, readable, on-brand, localized, formatted for the destination surface, and… Image generation is very popular and is even one of the most popular topics on this forum. As we know people enjoy experimenting with creating images and sharing them, thus those resources that could be used for GPT-5.4 are being used for the image generation. Again, just pure speculation; if others can provide verifiable facts related to “Selected model is at capacity” would be interested in reading them.