Discovered this workflow yesterday and had so much fun with it! Find the workflow at the end or learn how to make it yourself. I assume you have ComfyUI with Manager and IPAdapter installed, if not follow the tutorials of "Latent Vision" on YouTube and/or other sources Let's go!

🖼 Advanced Live Portrait 🔥 Jupyter Notebook + Tost AI + + ComfyUI 🥳 Thanks to Live Portrait Team ❤ github/PowerHouseMan ❤ 🧬code: github.com/PowerHouseMan/… 🍇runpod: github.com/camenduru/adva… 🗺workflow: github.com/PowerHouseMan/… 🍊jupyter: github.com/camenduru/Adva…

If you give FLUX1.1 a prompt like "IMG_1018.CR2" you get back images that are so very hard to tell they're AI. The realism here is kicked up a notch.

Wow a great tutorial by Mickmumpitz on creating consistent characters, including everything from using ComfyUI workflow to training a Flux LoRA using Fluxgym. Check it out!

Introducing ComfyUI V1, a packaged desktop application - Windows (NVIDIA), macOS (apple silicon), Linux - One click install for less technical users - Ships with ComfyUI manager - Auto-installed python environment Join the closed beta: comfy.org/waitlist

Trying the marble style on a hard fight scene with CogVideoX Fun Control + Reward LoRAs. You can get better results by stacking both LoRAs together. Made in ComfyUI with Jukka Seppänen custom nodes.

comfyUI的desktop发布了 下载地址：comfy.org/download 用户指南：comfyorg.notion.site/ComfyUI-Deskto… #comfyui

SpaceX has replaced Disney to make us dream! Thank you SpaceX

I let Claude computer use control Cursor to draw Mona Lisa in html canvas. Full code is in comments as images for computer use 🥳link to code also in comment I opened the html with live preview at some point manually so it can see what it is instructing Cursor to draw 🙂 after

Flux 1.1 Pro Ultra Raw is HERE. And it's the most realistic AI model I've ever seen! 10 examples (These are 100% AI) 👇 1.

Bob Ross

KLING TRY-ON → PRETTY SOLID Very satisfying. PROCESS: 01. Select "Virtual Try-On" 02. Upload a garment 03. Upload a model 04. Generate Image 05. Generate Video MIDJOURNEY BASE PROMPT: editorial photo, medium shot, asian female model with bleached blonde hair in a dynamic

1-Click HunyuanVideo ComfyUI Launcher This week we have tons of Pinokio updates coming up. Let's start the week with the best open source video AI we've ever seen: Hunyuan Video. A 1 Click launcher for ComfyUI-HunyuanVideoWrapper by Jukka Seppänen - run locally, for free.

OpenAI's Sora is a REALLY expensive slot machine with the occasionally incredible jackpot. In this video below we've got a $200/month AI model getting trounced by a free Chinese video model (Hunyuan). 🧵 A thread on reality vs. hype...

A lot going on here: - The main app runs on Python 3.12, Trellis is sandboxed in a conda environment running on Python 3.8. All seamless. No dependency hell - The 3D Preview is a standard component that can be used by any node that needs it. Lazy loaded only if needed

Gold tutorial on how to install HunyuanVideo on ComfyUI and use it for free!

AIで推しとの夢を叶えました！ Pika2.0 (Pika) made my dream come true with my fave character! #パーカーおじさん

First photo for ‘DUNE: PART 3’ from Denis Villeneuve Filming has now begun.

Coming soon, something very big is coming, Stanley Kubrick's 2001: A Space Odyssey in 8K, the master comes from a 70mm scan from the original camera negative, broadcast on NHK BS8K in Japan.