AVYD
ModelsPresetsStudioPricing

AVYD

The multi-model AI studio. Every model, one credit — on the web or inside ChatGPT.

One studio · one credit

Studio

Generate videoGenerate imagePresetsModelsPricing

AI video models

Veo 3.1Kling 3.0Wan 2.6Seedance 2.0Hailuo 2.3Sora 2

AI image models

Nano BananaFlux 2Seedream 5

Use cases

AI ad generatorTikTok video generatorProduct video generatorMusic video generator

© 2026 Avyd. AI video and image generation.

PrivacyTermsllms.txtsitemap.md
Models/Sora 2
Coming to AvydOpenAIvideo model

Sora 2

World simulation — and why Avyd routes those jobs elsewhere.

OpenAI announced it is retiring the Sora 2 API in September 2026, so Avyd is not building on it. World-simulation and physics-coherent workloads go to Veo 3.1 Pro and Kling 3.0 instead — both are live on the same credit system today.

Sora 2is in the Avyd integration pipeline. Create a free account and we'll email you at launch — members get first access, on the same credits as every other model.

  • Sora 2 pioneered long-shot world coherence and persistent physics
  • OpenAI is retiring the Sora 2 API in September 2026
  • Avyd routes world-simulation jobs to Veo 3.1 Pro and Kling 3.0
Spec sheet
Type
Text to video
Status
Not planned (API retired)
Strength
World simulation

Plans: Not planned — the Sora 2 API is being retired by OpenAI.

Credits: —

FAQ

Sora 2 on Avyd

Is Sora 2 available on Avyd today?

No, and it will not be. OpenAI announced it is retiring the Sora 2 API in September 2026. Avyd is not building on an API scheduled for shutdown. For world-simulation and physics-coherent shots, use Veo 3.1 Pro or Kling 3.0 — both are live now on the same credit system.

Why does this matter, and what is the Avyd alternative?

Model lineups change fast — exactly why one studio with swappable models beats locking into per-model subscriptions. When one provider retires an API, Avyd routes the workload elsewhere without any change to your workflow. Veo 3.1 Pro covers world-coherent long shots; Kling 3.0 handles photorealistic physics and action sequences.

More video models

Full catalog
LiveFlagship

Google DeepMind · video

Veo 3.1

Cinematic text-to-video with native audio, up to 4K.

Generate video Text to video
Coming soon

Kuaishou · video

Kling 3.0

The photorealism benchmark — people, fabric, and physics that hold up.

Get first access Text & image to video
Coming soon

Alibaba · video

Wan 2.6

Fast, expressive motion at a budget-friendly credit rate.

Get first access Text & image to video