gpt-image-2-vip / gpt-image-2-all: use the two OpenAI Images API endpoints for image generation
For alignment with the official calling standard and more stable supply, the two reverse models now recommend /v1/images/generations (text-to-image) and /v1/images/edits (image editing): same code and fully compatible size parameters as the official-relay gpt-image-2 — switching is just a model name change — and upstream resource supply for the Images API channel is more plentiful, so success rates are higher.
The chat endpoint (/v1/chat/completions) still works but is no longer recommended for image generation; it remains only for multi-turn iterative editing or passing online image URLs directly. The docs have been updated accordingly — see the GPT-Image-2-All overview and GPT-Image-2-VIP overview.
← Back to Live Updates · 📚 Monthly Archive