No.5472
im gonna kam to these pictures
No.5473
I'm going to train a monoe LoRa for Flux (hoping my card has enough VRAM), I'll upload it on Civit.AI when done.
No.5474
>>5473So far, so good.
First picture is prompt at start of training, second after ~800 steps.
No.5552
>>5551Jesus christ you should all be shot. Go back to twitter if you want to soy out over your faggy algorithms
(USER WAS FED TO AI MODEL FOR THIS POST) No.5561
>>5551ComfyUI is harder to master, but its shit tons of available nodes and the control it gave you over the models is unparalelled.
SD webui is perfectly fine for generating "simple" pictures, but once you step into complex productions, this is another story.
There's a lot of workflows available on the net you can use as bases.
Anyway, happy new year, hope this year we'll get even more tensor power and VRAM.
No.5577
>>5573It's another model based on SDXL and yes, Illustrious is quickly taking over Pony, a shit ton of LoRa is already available for it.
Personally, I've switched to Flux, Diffusion Transformers are way more powerful than U-net based models.
No.5578
>>5577Not like you need a LoRa anyway, at least for art styles. HassakuXL (the one I use) can imitate art styles of pretty much any somewhat popular artist.
> FluxIt still can't do anime style, isn't it?
No.5579
>>5578I was more referring to characters LoRa.
And Flux can do anime style image/characters without any problem, I've started training a few LoRas for specific characters (like the Monoe and Monoko ones I've posted earlier).
There's also some specific checkpoints of F1 dev that have been trained to produce only anime/manga pictures, pretty much like Pony and Illustrious.
No.5580
>>5579I've had no need of characters LoRa either, as the model I use recognizes all the (somewhat) popular characters too.
Good to know Flux can do anime style, but is it still requires more than 8 gigabytes of VRAM? If that's the case - guess I'll stick with Illustrious for now.
No.5581
>>5580Q4 quantized version of Flux Dev only needs around ~6.8GB of VRAM to run and produces pretty good results.
I've tested it on a mobile RTX 4060.
No.5629
>>5622Been a while, indeed.
I've been playing a lot with WAN 2.2 lately.
No.5631
>>5629And I've put RTX 3060 12 Gb into my server, so I can run LLM's and Stable Diffusion 24/7 and access them remotely. Also found new Illustrious models with better output quality.
I should try AI video generation someday… can this WAN thing run in SD?
No.5632
>>5631In SD? In Stable Diffusion? Really?
My dude, that's a whole different model, what are you even doing? Use ComfyUI or Pinokio.
No.5633
>>5631WAN 2.2 is usable through ComfyUI, I can share a simple working workflow so you'll only have to install ComfyUI Manager which will install all the needed nodes.
No.5634
>>5632I just got used to SD, that's why I'm still using it. ComfyUI doesn't seem to stand by it's name, at least to my liking. And Pinokio? What is Pinokio? Never heard of it before.
>>5633I'd appreciate if you could share it.
No.5635
>>5634ComfyUI is extremely powerful and let you do things you'll never be able to do with SD WebUI, but it's harder to use. I've been an early user of SD WebUI and I've switched to ComfyUI a "long" time ago.
And just look for the template called "Image to Video (New)" in the Templates section (filter by model and select WAN 2.2), it does pretty much the same thing as my workflow but uses a subgraph for a simplified view.
FP8 models of WAN 2.2 are around 9.5GB, they should run on your RTX 3060.
As for Pinokio, it's just a tool for managing installation of certains IA applications. There's no real need for it as ComfyUI has its own installer app on MacOS and Windows.
If you're on linux, just clone the repo like you probably did with SD WebUI. Create a python virtual env, install the dependencies with pip (use the requirements.txt file), then launch main.py.
