FramePack
Local image-to-video generator that predicts frames progressively so it runs on consumer GPUs.
These buttons open the developer's own site, repository or store listing in a new tab. wares.gg does not host downloads.
About FramePack
FramePack is the official implementation and desktop software for a research paper on next-frame prediction video diffusion. It generates video progressively, section by section, from a starting image and prompt.
Because it compresses the input context to a constant length, the workload does not grow with video length, which lets longer clips run on consumer graphics cards with limited memory. It runs as a Gradio interface in your browser.
Strengths
- Turns a still image into video on your own GPU
- Workload stays constant as the video gets longer
- Simple Gradio interface in the browser
Limitations
- Research code with infrequent updates
- Needs a capable GPU and a Python setup
- Generation is slow compared with hosted services
Details
- Pricing
- FreeFree and open source.
- License
- Apache-2.0
- Developer
- lllyasviel
- Platforms
- Windows, Linux
- How it runs
- Downloadable app
- Account
- Not required
- Works offline
- Yes
- Best suited for
- People with a gaming GPU who want to generate video locally
- Categories
- Video editing, Local AI tools, AI assistants
- Last verified
- Added
- Provenance
- Facts checked against the developer's own pages and store listings, 1 sources on file.
Alternatives to FramePack
Compare allSoftware that can replace FramePack for an important use case, and what changes if you switch.
Wan2GP
A local AI video generator built to run open video and image models on low-memory GPUs.
Wan2GP runs several open video models, plus image models like Flux, on low-memory GPUs from one app, but it is not open source.
Wan 2.2
Open, advanced large-scale video generation model for local use.
Wan 2.2 is an Apache-2.0 video model with variants sized to different GPUs, but it has no official GUI and is used through scripts.
LTX-2
Open local video model that generates synchronized audio and video.
LTX-2 generates synchronized audio and video together and includes a LoRA trainer, but its license is not a standard open license.
Mochi 1 (Genmo)
Open state-of-the-art video generation model from research lab Genmo.
Mochi 1 is an Apache-2.0 video model with consumer-GPU support through ComfyUI integration, with no repository activity since late 2025.
FramePack as an alternative
Listings that name FramePack as an alternative.
Hailuo AI
MiniMax's web app for creating AI videos and images from text prompts or photos.
FramePack is free, Apache-2.0 image-to-video software that runs locally on a gaming GPU, slower than hosted services and needing a Python setup.
PixVerse
An AI video generation platform that turns text prompts and images into video clips.
FramePack is free Apache-2.0 software that turns images into video locally on your own GPU, but needs Python setup and is slower than hosted services.
Similar software
Related functionality, not necessarily a direct replacement.
ComfyUI
Construct AI generation workflows with a visual node graph.
Pinokio
A desktop launcher that installs and runs open-source AI apps with one click.
Video2X
An open-source machine-learning tool for upscaling video resolution and interpolating extra frames.
Waifu2x-Extension-GUI
Upscales photos, GIFs and video with machine learning, and interpolates video frames.
nvitop
Interactive terminal viewer for NVIDIA GPUs and the processes running on them.
