Anyone else hate the new ComfyUI Login junk as much as me?

The way they are trying to turn the UI into a service is very off-putting to me. The new toolbar with the ever-present nag to login (starting with comfyui-frontend v 1.30.1 or so?) is like having a burr in my sock. The last freaking thing I want to do is phone home to Comfy or anyone else while doing offline gen.

Honestly, I now feel like it would be prudent to exhaustively search their code for needless data leakage and maybe start a privacy-focused fork whose only purpose is to combat and mitigate their changes. Am I overreacting, or do others also feel this way?

https://redd.it/1of0emr
@rStableDiffusion

Wan Animate - Tutorial & Workflow for full character swapping and face swapping
https://www.youtube.com/watch?v=2uYi_GsTV9A&lc=UgyDD6S-5dgKtXSQ3B14AaABAg

https://redd.it/1of9jk4
@rStableDiffusion

What is the best Anime Upscaler?

I am looking for the best Upscaler for watching Anime. I want to watch Rascal Does not Dream series, and was about to use Real-ESRGAN but its about 2 years old. What is the most best, and popular (ease of use) upscaler for anime?

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

Wan 2.2 T2I speed up settings?

I'm loving the output of wan 2.2 fp8 for static images.

I'm using a standard workflow with the lightning loras. 8 steps split equally between the 2 samplers gets me about 4 minutes per image on a 12GB 4080 at a 1024x512 res which makes it hard to iterate.

as I'm only interested in static images I'm a bit lost as to what are the latest settings/workflows to try speed up the generation?

https://redd.it/1ofbl9n
@rStableDiffusion

Pony V7 impressions thread.

I'm not going to leak the model, because that would be dishonest and immoral. It's supposedly coming out in a few hours.

Anyway, I tried it, and I just don't want to be mean. I feel like Pony V7 has already been beaten so bad already. But I can't lie. It's not great.

*Many of the niche concepts/NSFXXX understanding Pony v6 had is gone. The more niche, the less likely the base model is to know it

*Quality is...you'll see. lol. I really don't want to be an A-hole. You'll see.

*Render times are slightly shorter than Chroma

*Fingers, hands, and feet are often distorted

*Body horror is extremely common with multi-subject prompts.

https://preview.redd.it/kaqzwlcv06xf1.png?width=1024&format=png&auto=webp&s=eb990c3ddeca130b89b5d1d5de3e2d965cceab36



\^ "A realistic photograph of a woman in leather jeans and a blue shirt standing with her hands on her hips during a sunny day. She's standing outside of a courtyard beneath a blue sky."

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

Liquid Studios | Videoclip for We're all F*cked - Aliento de la Marea. First AI video we made... could use the feedback !
https://www.youtube.com/watch?v=uriGORyZzbA

https://redd.it/1off3km
@rStableDiffusion

The final generated image is the telos (the ultimate purpose).

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

Created AI product photography for perfumes and honestly shocked by the results

https://redd.it/1ofmna1
@rStableDiffusion

Its seems pony v7 is out
https://huggingface.co/purplesmartai/pony-v7-base/tree/main

https://redd.it/1ofp7zu
@rStableDiffusion

LTX 2 can generate 20 sec video at once with audio. They said they will open source model soon

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

🔥 Perplexity AI PRO - 1-Year Plan - Limited Time SUPER PROMO! 90% OFF!
https://redd.it/1ofrqcd
@rStableDiffusion

Texturing with SDXL-Lighting (4 step LoRA) in real time on RTX 4080

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

Built my dream AI rig.
https://redd.it/1ofntbt
@rStableDiffusion

What tools would you use to make morphing videos like this?

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