#ComfyUI workflows and more in our discord -- check them out --Batched IP adapter - QR code CN animation run using animate diff -8 image inputs - UDIO for music score #animatediff @jags111

Staged lora strength: 0, 0.25, 0.5, 0.75, 1.0, 1.25, 1.5 Morale of the story - you can go above 1.0 with FLUX perfectly fine

Visit Hedra's website. hedra.com/ Click "Stylize: Try to Transform your video with AI-powered effects." Click to add your image. Choose your desired style (e.g., Bobblehead, Anime, Kpop, Coconut, and more). Click "Apply Style" to transform your image. Choose an audio track to complement your stylized image. Click "Generate" to create your masterpiece. For more control, click "Try our full editor." Add custom audio or text (select a speaker for your text). Upload your photo. Select your style and click "Stylize." Apply changes and click "Generate." Save your final creation. #hedra #Ai @hedra_labs

Great overview tutorial! We’re working on some too

Animating yesterday's #FLUX woodcut Arnold using one of my favorite clips from the old soundboards This uses Follow-Your-Emoji / Reference UNet in ComfyUI, which did a better job than LivePortrait. Some comparison results in thread 🧵 #aivideo

Flux LoRA, 700 steps, 10 images

Good advice, railing.

WaterSplatting: Fast Underwater 3D Scene Reconstruction Using Gaussian Splatting arxiv.org/abs/2408.08206 Project: water-splatting.github.io/ Method ⬇️ 1 I 2

THIS SLAPS HARD

Thiel’s take that LLMs passing the Turing Test is a more meaningful moment in humans history than an achieving superintelligence is absolutely bonkers The association with Alan Turing gave it a status as a test that it just didn’t deserve

The weekly Radiance Field podcast with myself and @janusch_patas returns with @ruilong_li as our second guest! Ruilong needs almost no introduction as one of the core maintainers of @nerfstudioteam and is a research intern at NVIDIA. Next Thursday at 4pm EST/1pm PST right here! animation created with @LumaLabsAI

Added ps1 style flux v1.0 to FLUX LoRA the Explorer 🔎 huggingface.co/spaces/multim…

Built a better SD ultimate upscale in my node pack from scratch. Allows for MUCH faster rendering of slices, as well as doubles as a normal Ksampler when set to 1x1. Sliced cond. toggle for creative exploration using CNs, or simply as an upscaler. github.com/filliptm/ComfyUI_…

Online demos for BiRefNet on @huggingface Spaces! Is this the best background removal model out there? 🤯 MIT licensed. 5.5G GPU memory needed for inference for 1024x1024 images.🤩

» Flow « ✨👁️

Training FLUX LoRA's on @FAL is simple and you can do it yourself in 5 steps 👇

GROK side quest complete Back to our regular scheduled programming Playing with some new stuff 🤣 (not LivePortrait)

Check out this great work by @Undercover2009! We hear you on the aspect ratio!

Next level!

😜 Stay tuned for more sneak peeks—here’s a glimpse of what’s coming soon!