Sora API Shutdown — What Changed and What byesu Offers Today
byesu does not offer video generation right now
As of 2026-10-22 no video model is on sale on byesu — that includes Veo 3.1, Sora, Seedance, Runway and Kling. This page is an informational summary of the Sora API shutdown, not a migration offer. If video becomes available again, Video Generation will be updated.
OpenAI shut down the Sora API on 2026-09-24, as it had announced in advance. Under that announcement, requests to the Sora video endpoints from 2026-09-25 onward return HTTP 410 Gone, and OpenAI named no successor video API — integrations built on sora-2 or sora-2-pro had to move to another provider. byesu is an AI API gateway with OpenAI-compatible and Anthropic-native endpoints; today its media line-up is image models only (see below).
Shutdown timeline
| Date | What happened |
|---|---|
| Until 2026-09-24 | The Sora API was still available; this was the migration window |
| 2026-09-24 | Official shutdown date |
| From 2026-09-25 | Sora video endpoints return 410 Gone per OpenAI's announcement; no replacement inside OpenAI's API |
The whole Sora video surface was affected: task creation via POST /v1/videos, status polling, /content downloads, and any pipeline downstream of them. There was nothing to "upgrade to" within OpenAI's API, so this was a full provider migration rather than a version bump.
If you still need video generation
byesu cannot serve video at the moment, so a remaining Sora integration has to move to a video provider's own API. Whichever provider you choose, the same checklist applies:
- Inventory every call to the Sora endpoints (create, poll,
/contentdownload) and any webhook handlers. - Map parameters — clip length (
seconds), resolution (size) and reference-image inputs differ between providers; check the new provider's supported values. - Check the async flow — most video APIs still use create → poll → download, but status names, polling limits and result URLs vary.
- Re-test your prompt library side by side — identical wording can render very differently across video models.
- Plan for failures — decide how your app handles failed or timed-out tasks and how they are billed by the new provider.
What byesu offers today: image generation
Five image models are on sale, billed per image, with the same price at 1K / 2K / 4K (prices as of 2026-10-22):
| Model | Price |
|---|---|
gpt-image-2 | $0.009 / image |
gpt-image-2.5 | $0.009 / image |
gemini-3.0-pro-image-ultra | $0.009 / image |
gemini-nano-banana-2.1-ultra | $0.008 / image |
gemini-3.1-flash-image-ultra | $0.007 / image |
Token must use the media group
Image models require a token in the media / media-gen group, otherwise you'll get a "no available channel" error. Create one in the console before testing.
The same account also reaches text models — Claude (Opus 5.5 / Sonnet 5.5), GPT-6.1 Sol, Gemini 3.8 Flash, Grok 4.7 and more — billed pay-as-you-go, with top-ups via USDT or Alipay. Text prices are shown in the console after you sign in.
Related links
- Image / Video generation overview
- Image generation guide
- Image API pricing compared
- Quickstart (3-step setup)
- Online generation playground
- Console sign-up / create a token
FAQ
When did the Sora API shut down?
OpenAI shut down the Sora API on 2026-09-24, as it had announced in advance. Under that announcement, requests to the Sora video endpoints from 2026-09-25 onward return HTTP 410 Gone, and OpenAI named no successor video API — integrations built on sora-2 or sora-2-pro had to move to another provider.
Does byesu offer a Sora replacement or a video generation API?
Not right now. As of 2026-10-22 no video model is on sale on byesu — that includes Veo 3.1, Sora, Seedance, Runway and Kling. If video becomes available again, the Video Generation page in the docs will be updated.
What media models can I use on byesu today?
Image models only: gpt-image-2 and gpt-image-2.5 at $0.009 per image, gemini-3.0-pro-image-ultra at $0.009, gemini-nano-banana-2.1-ultra at $0.008 and gemini-3.1-flash-image-ultra at $0.007 per image (prices as of 2026-10-22), the same price at 1K / 2K / 4K. The same token setup also reaches text models such as Claude, GPT, Gemini and Grok.
Do I need a special token for the image models?
Yes. Image models live in the media / media-gen group, so create a token in that group in the byesu console; a token from another group returns a no available channel error.