OpenAI shut down the Sora API on September 24, 2026. Six entries went at once: the Videos API, sora-2, sora-2-pro and three dated snapshots. The recommended replacement column for all six rows still shows a dash. Most other deprecations on that page name a successor model. Sora has none, because OpenAI did not replace it. If anything in your stack still calls it, migration means changing vendors.
Last verified: 2026-10-20, against OpenAI's deprecations page and help center, Google's Gemini API pricing and deprecations pages, and Runway's API pricing. Updated after the shutdown date; the screenshots below were captured on 2026-09-19.
Looking for where to go after the Sora API shutdown? Start with our Sora alternative guide, then come back here for the deprecation timeline, export steps, and per-second cost comparison of the replacement APIs.
TL;DR
- Shutdown date: September 24, 2026, announced March 24, 2026. That date has passed; calls to the Videos API and Sora 2 models no longer work.
- The Sora web and app experiences went dark earlier, on April 26, 2026. The API was the second and final stage.
- No successor. The recommended replacement column is empty for all six entries, so every migration path leads to a different vendor.
- Check the target's own deadline. Google's Veo 3.1 preview models on the Gemini API shut down on October 22, 2026; Google points to Gemini Omni Flash, at roughly $0.10 per second of 720p video.
- Old reference prices:
sora-2was $0.10 per second at 720p, andsora-2-prowas $0.70 per second at 1080p. - Export what you can. OpenAI says Sora data is permanently deleted after discontinuation and after any final export window closes.
Quick Answer
If something still calls the Videos API, it has been failing since September 24, 2026; the endpoint and models are no longer accessible. OpenAI has not published a successor, so you are choosing between vendors rather than upgrading a model string. The realistic options are Google's video models (Gemini Omni Flash on the Gemini API, or the generally available Veo 3.1 models on Google Cloud), Runway's pay-as-you-go API, and Seeddance's video API. Google is the closest match on price and built-in audio, but do not start new work on the veo-3.1-*-preview IDs, which shut down on October 22, 2026. Runway is the easiest single invoice across several model families. Seeddance is the option to weigh when you need long reference-driven generation.
OpenAI API deprecations page, captured 2026-09-19. All six Sora rows list "---" under Recommended replacement.
The timeline, from OpenAI only
| Date | What happened | Source |
|---|---|---|
| March 24, 2026 | OpenAI notified developers using the Videos API and Sora 2 aliases and snapshots of their deprecation | Deprecations page |
| April 26, 2026 | Sora web and app experiences discontinued | Help center |
| September 24, 2026 | Sora API discontinued; Videos API and all Sora 2 models removed | Both |
| October 20, 2026 | Deprecations page still lists no recommended replacement for any Sora entry | Deprecations page |
OpenAI's stated policy is at least six months of notice before retiring a generally available model. March 24 to September 24 is exactly six months. Sora was not cut short, and it was not extended. OpenAI's video generation guide now marks the API as unavailable.
OpenAI help center, captured 2026-09-19, confirming both discontinuation dates and the export process.
Exactly what is being removed
Six entries, all removed on 2026-09-24:
- Videos API
sora-2sora-2-prosora-2-2025-10-06sora-2-2025-12-08sora-2-pro-2025-10-06
OpenAI describes the effect in general terms rather than naming a status code: at the time of shutdown, the model or endpoint is no longer accessible. Several secondary write-ups have circulated a specific HTTP code for what Sora requests return now. That figure is not on OpenAI's page, so treat it as unconfirmed and handle any failure from those calls generically rather than against one expected response.
Who this actually affects
API integrators. Anything calling the Videos API in production stopped working on September 24. If a job still calls it, it is failing now, and there is no workaround on OpenAI's side.
Teams with Sora in a workflow. Automations, internal tools and agent chains that shell out to Sora need a new provider wired in, plus new response parsing, because no alternative uses OpenAI's request shape.
Content teams with assets in Sora. OpenAI's help center still points to the Export button at sora.chatgpt.com/sunset, with an email when the archive is ready. It also says a final export window will only be offered if possible, with notice by email first, and that data associated with your use of Sora is permanently deleted after discontinuation and after any such window passes. If you have not exported yet, try now and do not assume more time. Purchased ChatGPT or Sora credits can still be used for Codex.
Migration options compared
| Option | API availability | Price per second of video (USD) | Max resolution | Audio | OpenAI-compatible |
|---|---|---|---|---|---|
sora-2 (removed) |
Ended 2026-09-24 | Was $0.10 at 720p; $0.05 batch | 720p | Yes | Native |
sora-2-pro (removed) |
Ended 2026-09-24 | Was $0.30 at 720p, $0.50 at 1024p, $0.70 at 1080p | 1080p | Yes | Native |
| Google Gemini Omni Flash | Gemini API, paid tier | Token-billed; about $0.10 per second at 720p | 720p native; 1080p and 4k upscaled | Yes | No |
| Google Veo 3.1 Standard (preview ID) | Gemini API until 2026-10-22 | $0.40 at 720p and 1080p; $0.60 at 4k | 4k | Included by default | No |
| Google Veo 3.1 Fast (preview ID) | Gemini API until 2026-10-22 | $0.10 at 720p; $0.12 at 1080p; $0.30 at 4k | 4k | Included by default | No |
| Google Veo 3.1 Lite (preview ID) | Gemini API until 2026-10-22 | $0.05 at 720p; $0.08 at 1080p | 1080p | Included by default | No |
| Runway API | Pay as you go, $0.01 per credit | Per model; Seedance 2.5 is $0.20 at 480p, $0.30 at 720p, $0.68 at 1080p | Varies by model | Varies by model | No |
| Seeddance API | Public REST API | Credit-based, not per second | 1080p on the Seedance 2.0 family, 720p on Seedance 2.5 | Per-model generate_audio support |
No |
Three honest notes on that table. Nothing on this list is OpenAI-compatible, so no migration is a base-URL swap; all of them are a rewrite of the request and polling code. Google offers no free tier for Veo 3.1 or Gemini Omni Flash, so there is no zero-cost way to evaluate either. And the three Veo 3.1 rows are the preview IDs on the Gemini API, which shut down on October 22, 2026; a migration finished after that date should target Gemini Omni Flash or the generally available Veo 3.1 models on Google Cloud.
Gemini API pricing page, captured 2026-09-19, showing per-second USD rates for all three Veo 3.1 variants. The rates were unchanged on 2026-10-20, when the page also listed the October 22 shutdown.
Google Veo 3.1 and Gemini Omni Flash
On the Gemini API, Veo 3.1 ships as three preview IDs: veo-3.1-generate-preview, veo-3.1-fast-generate-preview and veo-3.1-lite-generate-preview. Google's deprecations page gives all three a shutdown date of October 22, 2026 and names gemini-omni-1.1-flash as the replacement. Gemini Omni Flash became generally available on August 27, 2026. It generates video with audio, defaults to 720p, offers 1080p and 4k as upscaled output, and is billed per token, which Google estimates at about $0.10 per second of 720p video. The generally available Veo 3.1 models remain on Google Cloud's Gemini Enterprise Agent Platform.
Best for: teams moving off sora-2 or sora-2-pro who want the price to go down rather than up, and who need synchronized audio to keep working.
Not for: anyone who needs to test before committing budget, or anyone planning to build on the Veo 3.1 preview IDs. There is no free tier for Veo 3.1 or Gemini Omni Flash.
Runway API
Runway's developer tier is pay as you go: credits cost $0.01 each, and every model has its own rate in credits per second. Seedance 2.5, for example, costs 20, 30 or 68 credits per output second at 480p, 720p or 1080p, plus a charge for input and reference video and an 80-credit minimum per generation. A Model Router can pick across several labs' models and bills at the rate of whichever model it selects, so one integration reaches more than one provider.
Best for: teams that want to stop betting on a single provider after being burned by exactly that, and who value one invoice over the lowest unit price.
Not for: workloads where you need to pin one specific model's behaviour permanently, since routing across labs is the point of the product.
Runway pricing page, API tab, captured 2026-09-19. Rates in the text above were rechecked on 2026-10-20 in Runway's API pricing documentation.
Seeddance API
The Seeddance API creates a task with POST https://www.seeddance.io/v1/videos/generations, then polls GET /v1/tasks/{task_id}, with a bearer token from Settings and a separate GET /v1/credits for balance. Six public model IDs are documented, from seedance-1.0-pro-fast through seedance-2.5. The Seeddance API documentation covers the request shape and the polling loop.
Best for: reference-heavy generation. seedance-2.5 takes up to 30 images, 10 videos and 10 audio files as references, produces 4 to 30 second output, and lets prompts address assets directly as @image1, @video1 and @audio1.
Not for: a straight resolution match with sora-2-pro. See the boundary below before you pick it.
Seeddance API models documentation, captured 2026-09-19.
When to choose Seedance 2.5, and when not to
The useful version of this section is the part most vendor pages leave out.
Resolution depends on the route. On the Seeddance API, seedance-2.5 is documented at 480p to 720p, while the Seedance 2.0 family runs 480p to 1080p. So if you were calling sora-2-pro at 1080p, seedance-2.0 is the closer match on that API. The Seedance 2.5 model itself does reach 1080p through providers such as BytePlus ModelArk and Runway, which the Seedance 2.5 API guide compares.
Length and references are the win. Seedance 2.5 produces 4 to 30 second clips and accepts far more reference material than a text prompt plus one image. If your Sora usage was really "generate a shot that matches these existing assets," that is where it does better than a like-for-like swap.
The API is not OpenAI-compatible. Custom request schema, custom task polling. Budget for a rewrite, not a config change. Model and resolution access is also gated by account plan, and the API returns a 403 or 422 before creating a task if you ask for a combination your plan does not allow, which is at least a fast failure rather than a surprise bill.
Audio is per model. generate_audio only works with models that support synchronized audio. Sora 2's native audio is the feature most likely to have no direct equivalent, so confirm it for your chosen model before you assume parity.
If you want to evaluate output quality before touching the API, the Seedance 2.5 model page covers capabilities, and you can create a video from a prompt in the browser first. Credit costs per plan are on the SeedVideo AI pricing page.
What to do before September 24
That deadline has passed. If you are still cleaning up, this is the order that matters now:
- Grep your codebase for
sora-2,sora-2-proand the Videos API endpoint. Include scheduled jobs and agent tool definitions, which are the ones people forget. Those calls are failing, possibly quietly. - Try the export at
sora.chatgpt.com/sunsetnow if you have not already. Deletion is permanent once any final export window closes. - Pick a target on price and audio, not on brand, and check that target's own deprecation page. The Veo 3.1 preview IDs shut down on October 22, 2026.
- Rewrite the call and the polling loop. None of the alternatives accept OpenAI's request shape.
- Handle failure generically. OpenAI has not published the exact response Sora calls return after shutdown, so do not code against one specific status.
For a wider comparison than the four options here, the best Sora alternatives roundup covers the consumer-facing tools too, the AI video tool price comparison tracks rates across vendors, and the Seedance 2.5 benchmark data shows where it actually places on third-party leaderboards rather than on its own marketing.
FAQ
When exactly does the Sora API shut down?
It shut down on September 24, 2026. OpenAI notified developers on March 24, 2026, and lists the Videos API, sora-2, sora-2-pro, sora-2-2025-10-06, sora-2-2025-12-08 and sora-2-pro-2025-10-06 with a shutdown date of 2026-09-24. The Sora web and app experiences were already discontinued on April 26, 2026.
Is OpenAI releasing a replacement for the Sora API?
No. On OpenAI's deprecations page, the recommended replacement column is empty for all six Sora entries, shown as three dashes. Other deprecations on the same page name a successor model, which makes the blank column a deliberate signal rather than an oversight. There is no announced OpenAI video generation endpoint to migrate to.
What is the cheapest replacement for the Sora API?
On published rates, Google's Veo 3.1 Lite preview is $0.05 per second at 720p and $0.08 at 1080p with audio, but that ID shuts down on October 22, 2026. Its named replacement, Gemini Omni Flash, works out to about $0.10 per second at 720p, the same rate sora-2 charged at 720p. On Runway, each model has its own credit rate at $0.01 per credit. Google has no free tier for these models, so evaluation costs money from the first call.
Will my Sora videos be deleted?
Yes. OpenAI says that after Sora is discontinued, and after the period of any final export window passes, it permanently deletes data associated with your use of Sora. Its help center still describes exporting at sora.chatgpt.com/sunset with the Export button, followed by an email when the export is ready, and says users will be notified by email if a final export window is offered. If you have not exported, try it now.
Can I just change the base URL to another provider?
No. None of the available video generation APIs accept OpenAI's Videos API request shape. Veo uses Google's SDK through the Gemini API or Vertex AI, Runway uses its own platform API, and Seeddance uses a custom create-then-poll flow at /v1/videos/generations. Plan for rewriting the request construction and the polling logic rather than swapping a base URL.
Should I migrate to Seedance 2.5 or Seedance 2.0?
On the Seeddance API it depends on resolution. seedance-2.5 is documented at 480p to 720p there, while the Seedance 2.0 family runs 480p to 1080p. If you were generating at 1080p on sora-2-pro, seedance-2.0 is the closer match on that API. Choose 2.5 when you need long output of up to 30 seconds or heavy reference input of up to 30 images, 10 videos and 10 audio files.
What happens to my unused ChatGPT or Sora credits?
Per OpenAI's help center, any purchased ChatGPT or Sora credits can still be used for Codex. For refunds, OpenAI points to its standard ChatGPT subscription refund process rather than offering a Sora-specific one.
Start the migration this week
The shutdown date has passed, so any remaining Sora call is already broken. Search your repository first, try the export second, and pick a target third, checking that target's own shutdown dates before you commit. If reference-driven generation is what you were using Sora for, the Seedance 2.5 model page is the place to check capabilities and compare output in the browser before writing any integration code. For the request-level changes — endpoints, model IDs, pricing by resolution and status mapping — the Seedance 2.5 API guide walks through a full five-second call.
