Amuse AI Video Generator

Amuse is a local Windows studio from TensorStack, the app AMD partners with for on-device image and video. Official partner copy is short clips from a prompt, a sketch, or existing media — no account, no cloud job. Length follows the model you load. ImgToVid does not run Amuse; the generator on this page drafts on Seedance 2.5.

01 — Capabilities

A desktop app, not a hosted clip API

Amuse’s public pitch is privacy and hardware: CPU, GPU, or NPU on your machine. It does not publish one studio-wide duration the way a cloud model page would.

01
Runs on the PC in front of you
AMD’s partner page and TensorStack’s repo describe a free local install. EZ Mode is the no-setup path. Expert Mode lists 100+ models. You are not posting a job to a remote queue.
02
Short clips, model by model
Official wording is short videos and footage restyle, including ControlNet (Canny, OpenPose). TensorStack names video pipelines such as LTX, Wan 2.2, and CogVideoX. Those engines keep their own limits.
03
Still the main job; video is extra
Amuse started as a local image app. Video sits beside inpaint, upscale, interpolation, and a media gallery. AMD’s 3.0 launch copy called the video side draft-quality — not a 4K cloud take.
04
Audio tools are local too
The public repo lists Supertonic for voice and Whisper for speech recognition, plus ACE-Step for audio. That is optional local tooling, not a native soundtrack baked into every clip.

02 — Prompt to output

Prompts that match the clips above

Each prompt describes the picture on its card. “Use this prompt” sends that text to Create. The generator on this page runs Seedance 2.5, not Amuse.

01 · Portrait
PromptClose-up of a young man with curly hair in a grey knit sweater, soft window light. He glances away, then smiles. Slow push-in, shallow depth of field.
02 · Product
PromptA white leather sneaker on a pale grey studio backdrop, soft diffused light. The camera orbits slowly. Stitching stays sharp, the look stays minimal.
03 · Street
PromptA slow dolly down a cobblestone lane in an old European village. Climbing roses on stone walls, café tables in morning sun, a few people walking in the distance.
04 · Character
PromptA young woman with shoulder-length brown hair in an olive t-shirt, late-afternoon gold light. She turns her head and looks off. Breeze in her hair, a handheld feel.

03 — Where it fits

Jobs that stay on the machine

Amuse is for people who already have a GPU and do not want a cloud account. A talking-avatar campaign or a multi-scene agent is a different product.

A prompt you do not want to upload
Local generation is the documented reason to install Amuse: the still or the short clip never leaves the PC.
A clip you will restyle, not host
Official copy includes restyling existing footage with ControlNet. You are driving a local pipeline, not buying minutes of a presenter.
Trying LTX or Wan without a SaaS tab
If the experiment is which open video pipeline runs on your card, TensorStack’s named list is the public menu. Limits follow that pipeline.
A draft shot on ImgToVid
Write the shot here. Seedance 2.5 carries the prompt into Create. Membership and credits are checked there.

04 — Verdict

What Amuse actually is

Strengths

Free local Windows app; no login on the partner page
Text, sketch, or existing media into a short clip
Named pipelines include LTX, Wan 2.2, and CogVideoX
EZ Mode plus Expert Mode with 100+ models

Trade-offs

No Amuse-wide published clip length or frame size
A desktop install, not a cloud talking-avatar studio
Amuse is not available on ImgToVid
Marked coming soon on this site

05 — Side by side

Amuse, Veo 3.1 and Kling 4.0

Amuse is a local app that loads other models. Veo 3.1 and Kling 4.0 numbers below are from those vendors’ public docs.

	Amuse	Veo 3.1	Kling 4.0
Developer	TensorStack	Google DeepMind	Kuaishou
Single clip	Short; set by the local model	4, 6 or 8 seconds	3–30 seconds
Picture	Follows the model you load	720p, 1080p, 4K at 24 fps	720p, 1080p and 4K listed
Sound	Optional local voice/STT tools; no clip-wide spec	Native, with the picture	Stereo, with lip sync
Start from	Prompt, sketch, or local footage	Prompt, still, first and last frame	Keyframes or 15 references
On ImgToVid	Coming soon	Coming soon	Coming soon

06 — How it works

Start a clip on ImgToVid

Step 1
Describe one shot
Subject, action, camera. Amuse will not be the model that runs it.
Step 2
Use the generator on this page
It runs Seedance 2.5, because Amuse is not offered here. The prompt and the settings carry into Create.
Step 3
Generate in Create
Membership and credits are checked there. Coming back to this page does not switch the model over to Amuse.

07 — FAQ

Amuse, in plain terms

What is Amuse?

Amuse is TensorStack’s free local Windows app for images and short videos. AMD partners on the hardware path. You generate on your CPU, GPU, or NPU from a prompt, a sketch, or existing media.

How long is an Amuse clip?

Public pages say short videos. There is no Amuse-wide duration. Length follows the local pipeline you load — LTX, Wan 2.2, and CogVideoX are names on TensorStack’s repo.

Does Amuse generate audio?

The repo lists separate local tools: Supertonic for voice, Whisper for speech, ACE-Step for audio. Video generation itself does not publish a native-soundtrack toggle.

Do I need an account?

AMD’s partner copy says no login and no internet required for generation. That path is not available on ImgToVid.

Can I generate Amuse videos on ImgToVid?

No. The generator on this page runs Seedance 2.5 so you can draft a shot. Amuse is marked coming soon.

Can I use the results commercially?

Commercial use of anything you generate on ImgToVid depends on your plan. Check Pricing and our Terms before you publish, and make sure prompts and uploads do not infringe someone else’s rights.