Skip to content
Agents tracked: 492 Releases (7d): 545 Agent status: 1 with issuesSince last limit reset: Claude 9d · Codex 7d For AI agents: Markdown Updated Nov 20, 2026

Stable Video Diffusion (SVD-XT)

by Stability AI

Image-to-video model that turns one still frame into a short 25-frame clip at 576x1024.

AI video generators and platforms Open source

GitHub repoWebsite

2. Commercial use: huggingface.co

“this Agreement preserves free access to the Models for people or organizations generating annual revenue of less than US $1,000,000 (or local currency equivalent).”

Allowed after free registration for organizations under US $1M annual revenue, with the credit shown.

huggingface.co

License checked Oct 25, 2026 against 1 official page. Tap a verdict to see its source. How we checked Report an error

Stable Video Diffusion: turn one picture into a short clip

Stable Video Diffusion XT is a free, open model from Stability AI that turns one still picture into a short moving clip of a few seconds. You run it on a computer with a graphics card; there is no text prompt, the picture alone sets the scene. It is free for research and personal use, and for businesses under $1M a year after a free sign-up.

Where you can use it

Bring a product photo to life

Give a still product shot or illustration a few seconds of gentle motion for a page header or a social post.

More or less movement

Turn the amount of motion up for a lively clip or down for a calm loop, without any text instructions.

Clips on a mid-range graphics card

Make clips on a card with less than 8 GB of video memory by trading speed for memory, at the risk of slight flicker.

Motion for art and design

Animate artwork or concept frames for a portfolio, a music visual or a pitch, starting from images you made or own.

Moving storyboard previews Idea

Turn key frames of a storyboard into short moving previews, so a team sees the pacing before real filming or animation.

As reported by Stability AI: SVD-XT takes about 180 seconds per clip on an A100 80GB card with release defaults (SVD about 100 seconds); Trained to make 25 frames at 576x1024 from one frame of the same size.

Stable Video Diffusion (SVD-XT) popularity

44 / 100
Gid Score, same scale as developer tools
27,345
GitHub stars
Stable Video Diffusion (SVD-XT): GitHub starsStability-AI/generative-models 27,345

How the Gid Score is calculated

How to start

Tell your AI assistant

“Animate my image with stabilityai/stable-video-diffusion-img2vid-xt through diffusers (pinned version), resize it to 1024x576, use enable_model_cpu_offload and a fixed seed, export at 7 fps, and add "Powered by Stability AI" to the page that shows the clip.”

Or install it yourself.

pip install -q -U diffusers transformers accelerate
Show code

Load SVD-XT in half precision, resize your picture to 1024x576 and generate the frames.

frames = pipe(image, decode_chunk_size=8, generator=generator).frames[0]

Mistakes to avoid

pipe(image=image, prompt="a rocket taking off") with a text prompt
pipe(image, generator=generator) and steer motion with motion_bucket_id, fps and noise_aug_strength

The model takes no text; the picture alone sets the content.

Passing the photo at its original size
image = image.resize((1024, 576)) before generating

SVD-XT was trained on 576x1024 frames from a starting frame of the same size.

Running with the default decode_chunk_size on a small card and hitting out-of-memory
pipe.enable_model_cpu_offload(), pipe.unet.enable_forward_chunking() and a smaller decode_chunk_size

Video generation makes all frames at once, which takes much memory.

Shipping it in a commercial app without registration or credit
Register at stability.ai/community-license, show "Powered by Stability AI" and include the license and Notice file

The Community License requires all three for commercial use and distribution.

License

Stability AI Community License. Free for research and personal use. Businesses under $1M yearly revenue may use it commercially after a free sign-up and must show "Powered by Stability AI"; larger ones need a paid license.

Commercial use
With conditions

“this Agreement preserves free access to the Models for people or organizations generating annual revenue of less than US $1,000,000 (or local currency equivalent).” huggingface.co

Credit
Required

“prominently display “Powered by Stability AI” on a related website, user interface, blogpost, about page, or product documentation.” huggingface.co

Inside your app or game
With conditions

“If You are using or distributing the Stability AI Materials for a Commercial Purpose, You must register with Stability AI at (https://stability.ai/community-license).” huggingface.co

Pass the files on to your users
With conditions

Sharing the weights needs the agreement, the Notice file and the credit, and follows the same revenue limit.

“provide a copy of this Agreement to that third party” huggingface.co

Monetized YouTube
With conditions

“As between You and Stability AI, You own any outputs generated from the Models or Derivative Works to the extent permitted by applicable law.” huggingface.co

AI training
Not stated
Bots and bulk downloads
Not stated

AllowedWith conditionsNot allowedNot stated or not checked yet

Is Stable Video Diffusion (SVD-XT) free for commercial use? The full answer Not legal advice. Check the license page before you ship.

For your agent

License rules tell your agent what it may do with Stable Video Diffusion (SVD-XT). Paste them into your prompt or AGENTS.md.

License rules
- Stable Video Diffusion (SVD-XT) (AI video generators and platforms). Image-to-video model that turns one still frame into a short 25-frame clip at 576x1024. Price: free; commercial use with conditions; credit required; ship inside your app with conditions.
  Connect: download from https://huggingface.co/stabilityai/stable-video-diffusion-img2vid-xt (no API).
  Rule: Commercial use is free under $1M annual revenue; register with Stability AI first, above that get an enterprise license.
  Rule: Show "Powered by Stability AI" and ship the license and Notice file with any product that uses it.
  Rule: Follow Stability AI's Acceptable Use Policy; outputs may not be used to train other foundation models.
  License: https://huggingface.co/stabilityai/stable-video-diffusion-img2vid-xt/blob/main/LICENSE.md (checked 2026-10-25; details: https://agentgid.com/assets/stable-video-diffusion/)
Can an agent use it by itself?
No.
API, key or account
Not needed. Everything runs on your side.
MCP server
None found
Credit line, if you want one
Powered by Stability AI
This card for machines

Alternatives

Stable Video Diffusion (SVD-XT) This page

Free With conditions Download

Wan 2.2

Pick Wan 2.2 when you also want text-to-video and longer 480p or 720p clips from an open model family.

Free Commercial OK Download

LTX-2 (LTX-2.5)

Pick LTX-2 when the clip needs sound too; it makes video with audio from text, images or audio.

Free With conditions Download

Hypit

Pick Hypit when an AI assistant should remake whole short videos with captions and B-roll, using paid model services.

Free tier With conditions API, npm

Huobao Drama

Open-source platform that automates AI short-drama production from script and characters to storyboard and video compositing.

Free Non-commercial Download

CogVideo / CogVideoX

Open-source text-to-video and image-to-video generation models (CogVideo, CogVideoX) with inference and fine-tuning code.

Free With conditions Download

All ai video generators and platforms

Agents that do the same job

video editing agent skill, open source, 31.3K GitHub stars
AI video creation platform, open source, 22.2K GitHub stars
agent skill, open source, 22K GitHub stars
agent skill, open source, 21.2K GitHub stars
agent, open source, 6,968 GitHub stars

From AgentGid's agent catalog, not yet checked as asset sources: their license and price are on their agent pages.

Questions

Can I use it in a paid product?

Yes, if your revenue is under US $1M a year, you register with Stability AI and you show "Powered by Stability AI".

Who owns the clips I make?

You do, as far as the law allows, under the Community License.

Can I describe the motion in a text prompt?

No. Motion comes from the picture plus settings such as motion_bucket_id, fps and noise_aug_strength.

What is the difference between SVD and SVD-XT?

SVD makes 14 frames; SVD-XT is tuned further to make 25.

Is a personal hobby project free?

Yes. Research and non-commercial use is covered by the free license without registration.