Skip to main content
2026/10/9 10:47 (UTC+8) · Model Status · OpenAI ⚠️ gpt-image-2-vip and the gpt-image-2.5 -vip series have been adding capacity since yesterday and remain unstable The -vip series runs on reverse-engineered routes whose concurrency depends on the size of the account pool. Request volume rose sharply after the holiday and outpaced pool concurrency, so some periods see failed requests or slower queueing. We are continuously adding accounts to expand capacity. Current status by route:
  • gpt-image-2-vip, gpt-image-2.5-vip, gpt-image-2.5-sunburst-vip, gpt-image-2.5-flare-vip: scaling up, still fluctuating
  • Official gpt-image-2: ample concurrency
  • Official gpt-image-2.5-sunburst / gpt-image-2.5-flare: ample concurrency
For production workloads we still suggest keeping the official route as a fallback, along with client-side retries on failure. The official route is billed per token, unlike the per-image billing of -vip; see GPT-Image Official Overview. Thanks for your patience; we keep operating around the clock.
← Back to Live Updates · 📚 Monthly archive