1-Click INSTALL FRAMEPACK IMAGE-TO-VIDEO WebUI!
Added 2025-04-19 18:06:25 +0000 UTC!!! EDIT 20 APRIL: JUST UPLOADED V2 OF THE INSTALLER THAT NOW SHOULD SUPPORT 5000 SERIES GPU AND OLDER 2000/1000 SERIES AS WELL. JUST SELECT YOUR GPU SERIES AT THE BEGINNING AND THE INSTALLATION WILL DO ITS THING. BUT IF YOU SELECT OLDER GPU DO NOT CLICK ON THE UPDATE.BAT FILE AND JUST RUN THE RUN.BAT ONCE IT'S FINISHED INSTALLING!!! + ADDITION OF THE RUNPOD V2
Hey everyone! I've created a 1-click installer for the newest AI model tool called "FramePack" an amazing image to video webui that can generate up to 2 minutes of videos from a single image using only 6 gb of vram! It's just amazing!
You can check out the video right here: https://youtu.be/GywyMij88rY
The installer of course automates the entire install process for the webui!
LOCAL INSTALL
!!!You should have already installed the 12.6 CUDA TOOLKIT from here: https://developer.nvidia.com/cuda-12-6-0-download-archive or 12.8!!!
1. Download the FRAMEPACK-WEBUI_AUTO_INSTALL.bat
2. Run the bat file
3. Wait for the install to finish
4. ???
5. Profit 😎
IF YOU ARE USING RUNPOD:
1. Create an account if you haven't already: Runpod
2. Click on Pod (on the left side) then click deploy
3. Choose a GPU with at least 24gb of VRAM (a cheap 4090 or an A30 are great), then click modify template and choose the pytorch 2.1 template, then edit the template and choose 80gb for both the container and volume disk, then deploy on demand
4. Go to my pods, wait for everything to finish and then click "connect", then "Connect to Jupyter Lab [Port 8888]"
5. Then drag and drop the FRAMEPACK-WEBUI_AUTO_INSTALL-RUNPOD.sh file on the left side of the UI then click on the "Terminal" icon on the right side on the UI
6. Copy and paste these two lines then press enter:
chmod +x FRAMEPACK-WEBUI_AUTO_INSTALL-RUNPOD-V2.sh
./FRAMEPACK-WEBUI_AUTO_INSTALL-RUNPOD-V2.sh
7. Wait for everything to be installed
8. Click on the public url and have fun😎
IF YOU NEED TO STOP THE POD AND THEN RESTART IT. GO INSIDE THE FRAMEPACK FOLDER, CLICK ON THE TERMINAL ICON AND PASTE THESE TWO LINES THEN PRESS ENTER:
source venv/bin/activate
python demo_gradio.py --share
As always, supporting me on Patreon allows me to keep creating helpful resources like this for the community. Thank you for your support - now go have some fun😉!
Comments
This is great! Wondering if there are any workflows you have for adding lip synced audio to videos. 🤔
Rusty
2025-07-27 23:58:49 +0000 UTCHad to tinker around a while because I had issues with my 2070 Super. My solution; 1. Get a special repo for FramePack: https://github.com/freely-boss/FramePack-nv20 2. Head there, open a terminal, install requirements.txt python -m pip install -r requirements.txt 3. Install all dependencies (finding the correct ones was a pain in the butt) python -m pip uninstall -y torch torchvision torchaudio python -m pip install torch==2.7.0+cu126 torchvision==0.22.0+cu126 torchaudio==2.7.0 --extra-index-url https://download.pytorch.org/whl/cu126 python -m pip install --force-reinstall --no-deps xformers 4. Run demo_gradio.py python demo_gradio.py Make sure to be on Python 3.10 (google or chatgpt will help you get there) so it works.
Marc
2025-06-21 11:22:07 +0000 UTCSo I got everything to download, including the shaders, but no URL. It says press any key to continue, and it just closes down and doesn't open anything.
Joseph Daddona
2025-06-17 00:44:17 +0000 UTCSo I saw here to extend my page. I did do that but when I tried to run.bat again it said all was done but it doesn't open up anything and I see no URL. So I deleted everything to start again. When I run your program it no longer makes the frame pack folder. Just a zip file that can't be extracted. I've tried restarting my comp and using a new download of your program. Any help?
Joseph Daddona
2025-06-16 19:36:55 +0000 UTCI tried the new V2 on my 5090, but the file framepack_cu126_torch26.7z never gets created and there isn't a framepack folder after it all runs. But it says installation complete.
Tonic69
2025-05-22 07:31:26 +0000 UTCAny advice on this? What version of python do you need?
Colossal Pictures
2025-05-20 21:45:47 +0000 UTCyes
Aitrepreneur
2025-05-16 17:02:10 +0000 UTCWill there be a new studio version? The old one's a big loser.
SVT-Fashion-ART
2025-05-16 14:41:21 +0000 UTChello I am getting an error message when trying to install saying can not import from hugging face, any advice on this?
Colossal Pictures
2025-05-15 22:16:11 +0000 UTCsame
Martin Thoburn
2025-05-09 16:43:11 +0000 UTCIn run.exe Sage Attn is installed, so everything worked. However, checkpoint shards never loads beyond 0%, eventually leading to a "press any key to continue" notification. Web browser never launches. I have deleted and rerun everything about 6/7 times now and ot always gets hung up at that point. Can anyone help?
asdf
2025-05-07 20:14:41 +0000 UTCThis was a lot of fun to play with but I found every video after 10 seconds just kinda froze out as it had kinda exhausted your instruction. I would recommend peeps try this out: https://github.com/colinurbs/FramePack-Studio Exactly the same (in fact you can install it so it uses the same files you already have) but with 2 important changes I found useful: 1. You can create a timestamped prompt - i.e. it gives the video generator a new prompt over time timeline you set for the video. i.e. first 5 secs show a forest scene. At 10 secs: have a deer dart across the scene 2. It lets you include LORAs... you naughty, naughty people... It is pretty trivial to install if you don't have clashing dependencies... just follow the install on the repo. I did delete the old FramePack install from here to avoid confusion. But all that side still works as before.
Frunk
2025-05-07 10:13:42 +0000 UTC!! self.initialize_options() Traceback (most recent call last): File "/workspace/FramePack/SageAttention/setup.py", line 175, in setup( File "/workspace/FramePack/venv/lib/python3.10/site-packages/setuptools/__init__.py", line 117, in setup return distutils.core.setup(**attrs) File "/workspace/FramePack/venv/lib/python3.10/site-packages/setuptools/_distutils/core.py", line 186, in setup return run_commands(dist) File "/workspace/FramePack/venv/lib/python3.10/site-packages/setuptools/_distutils/core.py", line 202, in run_commands dist.run_commands() File "/workspace/FramePack/venv/lib/python3.10/site-packages/setuptools/_distutils/dist.py", line 1002, in run_commands self.run_command(cmd) File "/workspace/FramePack/venv/lib/python3.10/site-packages/setuptools/dist.py", line 1104, in run_command super().run_command(command) File "/workspace/FramePack/venv/lib/python3.10/site-packages/setuptools/_distutils/dist.py", line 1021, in run_command cmd_obj.run() File "/workspace/FramePack/venv/lib/python3.10/site-packages/setuptools/command/install.py", line 105, in run self.do_egg_install() File "/workspace/FramePack/venv/lib/python3.10/site-packages/setuptools/command/install.py", line 143, in do_egg_install raise NotImplementedError("Support for egg-based install has been removed.") NotImplementedError: Support for egg-based install has been removed. root@c7ec50184876:/workspace# can't install it. an error
Den Sh
2025-04-28 08:22:33 +0000 UTCAny idea why it keeps getting stuck? It was working fine and I didn't change anything except a new image. now it gets stuck on "moving dynamicswap_hunyuanvideotransformer3dmodelpacked to cuda:0 with preserved memory: 6GB 0% and GPU pegged at 100% / max VRAM
Don
2025-04-27 23:49:22 +0000 UTCDo you know how to resolve this error? ptxas C:\Users\User\AppData\Local\Temp\tmpv8tc40bd.ptx, line 895; error : Feature 'cvt with .f32.bf16' requires .target sm_80 or higher ptxas fatal : Ptx assembly aborted due to errors Repro command: D:\IA\framepack_cu126_torch26\system\python\lib\site-packages\triton\backends\nvidia\bin\ptxas.exe -lineinfo -v --gpu-name=sm_75 C:\Users\User\AppData\Local\Temp\tmpv8tc40bd.ptx -o C:\Users\User\AppData\Local\Temp\tmpv8tc40bd.ptx.o Unloaded DynamicSwap_LlamaModel as complete. Unloaded CLIPTextModel as complete. Unloaded SiglipVisionModel as complete. Unloaded AutoencoderKLHunyuanVideo as complete. Unloaded DynamicSwap_HunyuanVideoTransformer3DModelPacked as complete.
Claudinei
2025-04-26 01:24:03 +0000 UTCcan this be done with a AMD card?
Mol Jaa Boss
2025-04-25 20:51:02 +0000 UTC
