How come 4070 ti outperform 5060 ti in stable diffusion benchmarks by over 60% with only 12 GB VRAM. Is it because they are testing with a smaller model that could fit in a 12GB VRAM?
https://redd.it/1l85rxp
@rStableDiffusion

Self Forcing also works with LoRAs!

https://redd.it/1l87gjq
@rStableDiffusion

CAUGHT IN ANTARCTICA MERMAID OR SIREN?

https://redd.it/1l87fsc
@rStableDiffusion

Framepack Studio Major Update at 7:30pm ET - These are Demo Clips

https://redd.it/1l8dn58
@rStableDiffusion

Ultimate ComfyUI & SwarmUI on RunPod Tutorial with Addition RTX 5000 Series GPUs & 1-Click to Setup
https://www.youtube.com/watch?v=R02kPf9Y3_w&sdforall

https://redd.it/1l8eel0
@rStableDiffusion

FramePack Studio 0.4 has released!

This one has been a long time coming. I never expected it to be this large but one thing lead to another and here we are. If you have any issues updating please let us know in the discord!

https://github.com/colinurbs/FramePack-Studio

Release Notes:
6-10-2025 Version 0.4

This is a big one both in terms of features and what it means for FPS’s development. This project started as just me but is now truly developed by a team of talented people. The size and scope of this update is a reflection of that team and its diverse skillsets. I’m immensely grateful for their work and very excited about what the future holds.

Features:

Video generation types for extending existing videos including Video Extension, Video Extension w/ Endframe and F1 Video Extension
Post processing toolbox with upscaling, frame interpolation, frame extraction, looping and filters
Queue improvements including import/export and resumption
Preset system for saving generation parameters
Ability to override system prompt
Custom startup model and presets
More robust metadata system
Improved UI

Bug Fixes:

Parameters not loading from imported metadata
Issues with the preview windows not updating
Job cancellation issues
Issue saving and loading loras when using metadata files
Error thrown when other files were added to the outputs folder
Importing json wasn’t selecting the generation type
Error causing loras not to be selectable if only one was present
Fixed tabs being hidden on small screens
Settings auto-save
Temp folder cleanup

How to install the update:

Method 1: Nuts and Bolts

If you are running the original installation from github, it should be easy.

Go into the folder where FramePack-Studio is installed.
Be sure FPS (FramePack Studio) isn’t running
Run the update.bat

This will take a while. First it will update the code files, then it will read the requirements and add those to your system.

When it’s done use the run.bat

That’s it. That should be the update for the original github install.

Method 2: The ‘Single Installer’

For those using the installation with a separate webgui and system folder:

Be sure FPS isn’t running
Go into the folder where update_main.bat, update_dep.bat are
Run the update\_main.bat for all the code
Run the update_dep.bat for all the dependencies
Then either run.bat or run\_main.bat

That’s it’s for the single installer.

Method 3: Pinokio

If you already have Pinokio and FramePack Studio installed:

Click the folder icon in the FramePack Studio listed on your Pinokio home page
Click Update on the left side bar

Special Thanks:

RT_Borg https://github.com/RT-Borg
Anchorite [https://github.com/ai-anchorite](https://github.com/ai-anchorite)
Xipomus https://github.com/Xipomus
ptfq [https://github.com/pftq](https://github.com/pftq)
And thank you to everyone who has submitted a PR, feature request or bug, supported on Patreon, or just hung out in the Discord!

https://redd.it/1l8dtyl
@rStableDiffusion

Hey everyone back again with Flux versions of my Retro Sci-Fi and Fantasy Loras! Download links in description!

https://redd.it/1l8cqdc
@rStableDiffusion

Work for Artists interested in fixing AI art?

It seems to me that there's an untapped (potentially) market for digital artists to clean up AI art. Are there any resources or places for artists willing to do this job to post their availability? I'm curious because I'm a professional digital artist who can do anime style pretty easily and would be totally comfortable cleaning up or modifying AI art for clients.

Any thoughts or suggestions on this, or where a marketplace might be for this?

https://redd.it/1l8au3q
@rStableDiffusion

BUNRAKU - Trailer

https://redd.it/1l8m51k
@rStableDiffusion

As someone who is already able to do 3d modelling, texturing, animation all on my own, is there any new ai software that i can make use of to speed up my workflow or improve the quality of my outputs?

I mainly do simple animations of characters and advertisements for work.
For example, maybe if i am going through a mindblock i would just generate random images in comfyui to spark concepts or ideas.
But im trying to see if there is anything in the 3d side perhaps generate rough 3d environments from an image?
Or something that can apply a style onto a base animation that i have done up?
Or an auto uv-unwrapper?


https://redd.it/1l8n1p3
@rStableDiffusion

If you're out of the loop here is a friendly reminder that every 4 days a new Chroma checkpoint is released

https://redd.it/1l8r8av
@rStableDiffusion