Video generation

Stable Video Diffusion (SVD-XT)

Name: Stable Video Diffusion (SVD-XT)
Author: Stability AI

An image-to-video latent diffusion model that animates a single still image into a short clip at 576x1024. The XT variant is a fine-tune of the original 14-frame model and produces 25 frames.

A video generation model by Stability AI · read 2026-08-14 · sourcing medium

Resolution: 576x1024
Weights: Downloadable
API: No
Commercial use: Permitted subject to conditions

It was the first mainstream open-weights image-to-video model and is now historically important rather than competitive.

Commercial use — Permitted subject to conditions

Hugging Face license tag is 'stable-video-diffusion-community'.

Revenue threshold: free below $1,000,000 in annual revenue, chargeable above it.

Free only below a revenue threshold. Stability's Community License is free for research, non-commercial use, and commercial use by individuals or organisations with under USD $1M in annual revenue; above $1M an Enterprise License is required

The conditions above were extracted from the licence notes recorded during research, and each one shows the sentence it came from so you can check the reading. Read the licence before you rely on any of this. Nothing on this page is legal advice, and this site does not tell you a model is cleared for commercial use.

Specification

Vendor figures, read 2026-08-14Read from Stability AI's own documentation on 2026-08-14. Vendors change these between model versions without announcing it, so check theirs before you commit a shot to a number. Sourcing for this entry is mixed. Some fields came from pages we retrieved, others could not be confirmed and are left blank. Check the vendor's current documentation

What it takes in: image
What it makes: video
Maximum duration: not established
Maximum resolution: 576x1024
Aspect ratios: not established
Controls: input image, motion bucket id, conditioning augmentation / noise, fps conditioning id, decode chunk size, seed
Weights: Downloadable
API: No
Pricing: Free open weights under Stability's community licensing. Self-hostable on consumer hardware - it is by far the lightest model in this list, runnable in roughly 8-16GB VRAM with diffusers, though exact figures are not published on the model card. Stability's hosted platform API previously offered image-to-video; treat first-party API availability as unconfirmed, and Replicate/fal hosting is third-party.
Released or updated: 2023-11-21

Maximum duration and supported aspect ratios are not established here. That means no page we could retrieve stated them, not that the model lacks them.

Strengths

Lightest hardware requirement of any model here
Very well supported in diffusers, ComfyUI and Automatic1111 ecosystems
Free commercial use under $1M revenue
Simple, predictable motion controls (motion bucket)

Weaknesses

$1M revenue threshold is the strictest in this list
Only 25 frames at 576x1024 - roughly 1-4 seconds depending on playback fps
No text prompting at all; image conditioning only
Motion is minimal and often drifts
License history is muddled (research-only then community)

Best for

Lightweight local image animation, subtle motion on stills, and legacy pipelines already built around diffusers.

Where the sources disagree

Recorded rather than resolved. Picking the tidier number would hide the fact that the vendor's own pages do not agree.

The stability.ai/license page enumerates image, audio and 3D models but did not explicitly list Stable Video Diffusion in the fetched content, while the SVD model card points to that page for commercial terms. The $1M threshold is therefore inferred from the Community License it links to rather than from an SVD-specific clause. Also, no output fps is stated on the card, so duration cannot be computed.

What to understand before using it

Explainers naming the mechanisms this model's own documentation mentions.

Latent space: working at proxy scale names “latent diffusion”
Conditioning and control: everything you hand the model besides noise names “conditioning”
Text encoders and conditioning: how a prompt steers the picture names “conditioning”
Diffusion models: generation by denoising names “diffusion”

MCP servers that front Stability AI

Servers in the media MCP directory whose own description or homepage names this vendor. Absence of a server here is not evidence that none exists: most publish no tool list at all.

Audio Analyzer Audio and voice
Gives LLMs ears. Spectral, harmonic, rhythm, stereo, and structural audio analysis.
Audiolla Music
Self-hosted MCP server: audio stems, mastering, MIR analysis, DSP, MIDI, speech tools.
Civitai Image generation
Civitai MCP.
Cochlea Audio and voice
Render, analyze, and verify audio (WAV/FLAC/mp3/ogg) offline and deterministically via MCP tools.
Comfyui Image generation
MCP server + Claude Code plugin for ComfyUI: run workflows, generate images, manage models & VRAM.
Corent Image generation
Generate images and videos via the Corent API, auto model routing with provider fallback.

1 further server name Stability AI with lower confidence and are not listed, because a server naming a vendor may be using it for text rather than media.

See it from the vendor

Shotcraft hosts no sample output. Judging a model means watching its own showreel with your own eyes, and a clip re-hosted here would be a clip we chose.

Nearby in the directory

Adobe Firefly Video Model

The Adobe Firefly Video Model generates video from a text prompt, optionally guided by an image used as the first or final frame. Adobe's Firefly Serv

Video generation

Allegro

A small text-to-video model comprising a 2.8B VideoDiT and a 175M VideoVAE, released under Apache-2.0 in October 2024. It generates 88-frame clips at

Video generation

Amazon Nova Reel 1.1

Amazon Nova Reel generates video from text prompts or from text plus an image, and is available only through Amazon Bedrock. AWS documents output at 1

Video generation

Sources

Read on 2026-08-14. A source that has changed since then will not be reflected here until the next pass.