Sora 2 migration guide
Sora 2 is discontinued — what closed, when, and where its strengths live today
OpenAI closed the Sora app on April 26, 2026 and the Sora API on September 24, 2026, so Sora 2 can no longer be generated on Imgo or anywhere else. This page is now a migration guide: it maps each Sora 2 strength — cinematic realism with native audio, fast drafts, high volume, image references — onto the live video models that can take over that work today.
Capability snapshot
4K
final tier on Seedance 2.0
15s
per clip on live models
audio
native, on by default
See Sora 2 migration guide in action
A 27-second walkthrough of the workflow, the key benefits, and the results.
Core benefits
Practical value mapped to real production workflows.
Benefit
01
Cinematic realism + native audio → Seedance 2.0
Sora 2's signature was photoreal motion with its soundtrack baked in. That combination lives on Seedance 2.0 today: native audio on by default, output up to 4K, 5–15 second clips, and references to steer the look — up to 4 reference images, 3 reference videos or 3 reference audio tracks per task.
Benefit
02
Fast drafts → Seedance 2 Fast
Sora 2's quick iteration loop maps to Seedance 2 Fast: the same inputs and audio controls at up to 720p, at roughly 55% of Seedance 2.0's per-second rate (estimated). Draft the shot in Fast, then re-render the keeper on Seedance 2.0.
Benefit
03
High volume → Seedance 2 Mini
If Sora 2 was your volume engine, Seedance 2 Mini is the live equivalent: 4–15 second clips at up to 720p at roughly half of Seedance 2.0's rate (estimated), built for batches of social variants. Veo 3.1 and Kling 3.0 are pending on Imgo — listed as coming, never as available.
Recommended workflow
A simple three-step path from source material to a reusable result.
Step 1
Map each strength to a live model
Cinematic shots with audio go to Seedance 2.0, drafts to Seedance 2 Fast, volume batches to Seedance 2 Mini. The live model list shows what is generatable today; models not yet in the catalog are marked as coming.
Step 2
Re-shape your Sora prompts
Model phrasing does not transfer verbatim. Imgo's Prompt Builder restructures a Sora prompt into its Five Pillars format — subject, motion, camera, environment, style — and the Seedance adapter rewrites the result in Seedance phrasing.
Step 3
Re-render and compare
Run the same brief on the replacement model and compare takes with your archived Sora output before switching the pipeline over. Characters, brand anchors and references carry across unchanged.
Frequently asked questions
Is Sora 2 still available anywhere?
No. OpenAI closed the Sora app on April 26, 2026 and the Sora API on September 24, 2026. No platform — Imgo included — can generate Sora 2 clips anymore; files you already downloaded remain yours.
What is the closest live model on Imgo?
Seedance 2.0 is the closest match for Sora 2's cinematic realism with native audio. The same family covers the rest: Seedance 2 Fast for cheap drafts and Seedance 2 Mini for volume. Veo 3.1 and Kling 3.0 are pending — their own pages show availability pending until the live catalog lists them.
Do my old Sora 2 prompts work elsewhere?
Not verbatim — each model reads phrasing differently. Paste a Sora prompt into Imgo's Prompt Builder, keep the Five Pillars structure it extracts, and export through the Seedance adapter to get the same intent in Seedance phrasing, ready for the Video Generator.
Do my characters and references survive the move?
Yes. Imgo's character library and brand anchors are model-agnostic, and the live Seedance family accepts up to 4 reference images or 3 reference videos per task, so the same identity and style cues follow your prompts to the new model.
Next step
Want to generate today?
Browse the models that are live on Imgo right now, compare their settings and credit quotes, and start from one of them.