@linoy_tsaban Profile picture

Linoy Tsaban🎗️

@linoy_tsaban

Exploring the world of AI Art as a ML engineer @HuggingFace 🤗 | ✡️ & 🇮🇱 #BringThemHome 🎗️

Similar User
Reka photo

@RekaAILabs

Adina Yakup photo

@AdinaYakup

Ilyas Moutawwakil photo

@IlysMoutawwakil

ChunTe Lee photo

@lee_chunte

lunarflu photo

@lunarflu1

Jingya Huang photo

@Jhuaplin

Brigitte 🤗 photo

@BrigitteTousi

Omar Sanseviero photo

@osanseviero

Niels Rogge photo

@NielsRogge

Yacine Jernite photo

@YJernite

João Gante photo

@joao_gante

dylan photo

@dylan_ebert_

Yuvi photo

@yvrjsharma

Alara Dirik photo

@alaradirik

hysts photo

@hysts12321

Pinned

Introducing LEDTIS 🚀🎨 An intuitive editing approach for real images, combining DDPM inversion with semantic guidance 💫 @multimodalart 🤗 demo: huggingface.co/spaces/editing… 🤗 project page: editing-images-project.hf.space/index.html 🖌️ arXiv: arxiv.org/abs/2307.00522

LEDITS: Real Image Editing with DDPM Inversion and Semantic Guidance paper page: huggingface.co/papers/2307.00… demo: huggingface.co/spaces/editing… Recent large-scale text-guided diffusion models provide powerful image-generation capabilities. Currently, a significant effort is given to…



Linoy Tsaban🎗️ Reposted

MagicQuill: An Intelligent Interactive Image Editing System abs: arxiv.org/abs/2411.09703 MagicQuill @Gradio demo is now available on @huggingface Spaces space: huggingface.co/spaces/AI4Edit…

MagicQuill An Intelligent Interactive Image Editing System



Linoy Tsaban🎗️ Reposted

New Open Source SOTA Background Removal Model by BRIA AI

Tweet Image 1

Linoy Tsaban🎗️ Reposted

🛑 Someone didn't want you to see this post and coordinated a spam bot attack to limit its visibility—and it worked. More details below. 🛑 This is ReCapture, our Google project that obtains SoTA results on the previously unassailable problem of re-angling user-provided video.🧵


Linoy Tsaban🎗️ Reposted

Open-source ftw. Flux Dev LoRA by @pathmayur (AKA MindlyWorks!) DimensionX LoRA by @an_epsilon0 ComfyUI Implementation by @Kijaidesign

magical tbh



Linoy Tsaban🎗️ Reposted

It's time to share something exciting that we've been working on at Google. ReCapture is a method to generate new versions of a user-provided video with customized camera trajectories. It's basically post-capture cinematography for all. Project links are below. 🧵


Linoy Tsaban🎗️ Reposted

The community has uploaded more than 7000 Flux[dev] LoRAs to @huggingface 🤗🎊 Browse them all 🔍 and test them out for free 🖼️ ✨ ▶️ huggingface.co/models?other=b…


Linoy Tsaban🎗️ Reposted

✨ PD12M ✨, a 12.4 million high quality image-caption dataset for AI training 🎛️, featuring: - 🤖✏️ Florence-2 synthetic captions - 🌸 Aesthetic and safety filtered from 34M superset - 🔓 only public domain images superb release by @spawning_ huggingface.co/datasets/Spawn…


Linoy Tsaban🎗️ Reposted

aand Stable Diffusion 3.5 Medium is just out ✨ @stabilityai is rolling 🌀 A supercharged smol 🤏 2B model 🔋 with a permissive commercial license ▶️ Play with it now: huggingface.co/spaces/stabili…

Tweet Image 1

Linoy Tsaban🎗️ Reposted

IC-Light v2 was just released by @lvminzhang 🔦, now runs on FLUX, and it is the best relighting tool in the world 🌐, just like that Try out the official demo ✨📣 huggingface.co/spaces/lllyasv…

Tweet Image 1
Tweet Image 2

Yesss super cool & clever use/extension of existing techniques and pipelines 👾

We are making a bet that in the future most games will be entirely generated, and that generative game engines allow for things that traditional game engines do not. We know that LLMs exhibit emergent behavior, and we think this translates to games. link: generative-infinite-game.github.io



Linoy Tsaban🎗️ Reposted

Added a blog post with images from skipping each block in SD3.5-large. Hopefully this helps in choosing more efficient blocks to target during fine tuning. ostris.com/2024/10/23/ski…


good tip to maximize the benefits of🤏 quantized models:

So, when merging LoRA params into a 4bit quantized model, we should 1. dequantize the base model 2. merge the LoRA params into it 3. quantize it again See the difference (left is better).

Tweet Image 1


Well well well, Stable Diffusion 3.5 Large just landed 🛬 & it's already fully integrated in 🧨diffusers🔥 ▶️ Heres everything you need to know: huggingface.co/blog/sd3-5


Linoy Tsaban🎗️ Reposted

Guess who's back? Back again! 🎵 @StabilityAI is back, tell a friend 🎤 Stable Diffusion 3.5 Large is here 🔥 - 🏋️ 8B parameters - Full 💪 and 🏎️💨 4-step Turbo variant - 🧾 🤝 commercial use (for orgs below 1M year/rev) - 🧨 day-0 LoRA fine-tuning support

Tweet Image 1

Loading...

Something went wrong.


Something went wrong.