跳转到主要内容
2026/4/27 11:24 (UTC+8) · Model Status · OpenAI ⚠️ gpt-image-2 429 / 408 errors notice — The Default group’s gpt-image-2 official-relay capacity is tight, which can surface as 429 (rate-limit) or 408 (timeout). Available options:
  1. Retry: wait for capacity to free up.
  2. Switch token group: edit your token and move it to image2Enterprise (1.2x rate) — still official relay, suitable as a temporary bridge.
  3. Switch model: use the per-call gpt-image-2-all (reverse); that channel also has tight concurrency right now, ETA ~30 minutes to replenish.
📖 Official relay vs. reverse: /en/api-capabilities/gpt-image-2/vs-gpt-image-2-all
Back to Live Updates · 📚 Monthly archive