gpt-image-2-vip capacity restored · Images API remains the recommended format
Following up on the 7/6 service notice: upstream capacity for /v1/images/generations (text-to-image) and /v1/images/edits (image editing) has improved, and gpt-image-2-vip’s call success rate has risen accordingly. Same code and size parameters as official-relay gpt-image-2 — switch by changing the model name.
The chat endpoint (/v1/chat/completions) remains resource-constrained and isn’t recommended for image generation.
Thanks for your patience — we continue tracking upstream capacity.
← Back to Live Updates · 📚 Browse Archive