I feel like CC is not really devoting enough effort with this pipeline. What I can do in ComfyUI far exceeds the methodology currently be used in CC. In addition the current pipeline plugin seems very dated. WAN 22 is the current model. Not WAN 21. The Current NVIDIA card is the 50 series. The install people use VERY often is standalone PYTHON install for 40 or 50. ( I use 50). Yet all the info for CC plugin is outdated as far as I can see. This pipeline should WORK. Without issue. Particularly involving speed issues. I do 1080 x 800 5 secs in 5 minutes. Depending on LORAS used or complexity of AI continuity, consistency and quality. I can output things that would be unrecognizable from actual film or even the actors. CC still has a place in the pipelines but not if CC cannot catch up to what is going on. AI is GOING change EVERYTHING in the way these pipelines work. In essence I could create an entire movie in 6 months, no actors, no sets, no green screen…etc.
One more thing, with many AI models I can create characters are indistinguishable from reality. And can create all the movements via AI, not animation or skeletons, etc. ( Although I edit via NUKE or DaVinci, etc.)
CC needs to take this revolution seriously. This methodology is starting to feel dated. It may take a few more years, but change is coming.
Not sure to what area you are refereeing, but this is the latest on AI that I find.
(see below)
This is all outdated… If you have a better link, please let me know I will review let you know how relevant it is. I don’t think there is a lot of understanding about AI pipelines for, or by RL or CC.
Thanks
[Welcome to AI Render Open Beta – Start Here!]( discussionsDOTreallusionDOTcom official-welcome-to-ai-render-open-beta-start-here/14439
OUTDATED …way outdated 8/2025 latest. (ComfyUI portable Python?)
Graphics Card: NVIDIA 30 series 8GB (minimum), supported
WAN 2.1
Default models included: Stable Diffusion 1.5 for image generation, Wan2.1 Fun 1.3B Control, and Wan2.1 VACE 1.3B for video generation.
You don’t need to install all models in advance. Instead, simply download the ones required for the style presets you want to use from the AI Render Settings, ComfyUI tab.
@graphyx99 To me it looks like the previous ComfyUI workflows were some kind of tests by Reallusion to check out what is possible, but in their 2026 roadmap it looks like they will concentrate on offering some kind of proprietary cloud rendering service in the future and no more ComfyUI updates.
Cloud rendering service = censorship. So that is the totally wrong direction to go for artistic freedom. Just make a working plugin for the tried and true. Don’t reinvent the wheel.
In any event within 10 years or less, this type of CC creation for most things, will be ancient history.
I am trying to figure out what exactly your point is. The AI Render was an open beta, it was never promoted as a final product. It doen’t hurt anybody to keep it available and there have been several webinars on the subject of running it with any external ComfyUI configuration you want. It just provides an interface and you can do what you want with it.
It’s not being developed anymore, so the point is moot. Don’t you think RL should say that? Can’t do anything serious with an outdated workflow now can yah.