Log inSign up
Robin Rombach
756 posts
Image
user avatar
Robin Rombach
@robrombach
Krawallkrümel. Generative Models at bfl.ai, made with ❤️
Black Forest
Joined July 2019
563
Following
13.6K
Followers
  • Pinned
    user avatar
    Robin Rombach
    @robrombach
    Aug 1, 2024
    🔥 I am so damn excited to announce the launch of Black Forest Labs. We set ourselves on a mission to advance state-of-the-art, high-quality generative deep learning models for images and video, and make them available to the broadest audience possible. Today, we release FLUX.1
    user avatar
    Black Forest Labs
    @bfl_ai
    Aug 1, 2024
    We are excited to announce the launch of Black Forest Labs. Our mission is to develop and advance state-of-the-art generative deep learning models for media and to push the boundaries of creativity, efficiency and diversity.
    Image
    00:00
    388K
  • user avatar
    Robin Rombach
    @robrombach
    Feb 11, 2025
    Europe ♥️ Open Source.
    Image
    115K
  • user avatar
    Robin Rombach
    @robrombach
    Nov 21, 2023
    Our image-to-video model is finally out! I am especially excited about its finetuning capabilities for multi-view synthesis, which we further investigate in our paper. Check it out here: stability.ai/research/stabl… No SDS, just diffusion:
    Image
    00:00
    201K
  • user avatar
    Robin Rombach
    @robrombach
    Nov 28, 2023
    Excited to share our new work on real-time image synthesis: Adversarial Diffusion Distillation. We distill SDXL into a single-step model which still produces crisp samples. Weights: huggingface.co/stabilityai/sd… Paper: stability.ai/research/adver… Adversarial training where you need it.
    Image
    98K
  • user avatar
    Robin Rombach
    @robrombach
    Nov 11, 2025
    upgrade incoming.
    Image
    196K
  • user avatar
    Robin Rombach
    @robrombach
    Mar 19, 2024
    What's better than Adversarial Diffusion Distillation (ADD)? Yes: *Latent* ADD. We turn the diffusion transformer in SD3 into a discriminator, move distillation to latent space and get SD3-Turbo, which achieves SD3's quality in only four unguided steps. >arxiv.org/abs/2403.12015
    Image
    21K
  • user avatar
    Robin Rombach
    @robrombach
    Aug 10, 2022
    The weights of Stable Diffusion, a latent text-to-image diffusion model, are open for academic research upon request! See github.com/CompVis/stable… for details how to access the model and the model card.
    Image
    Image
  • user avatar
    Robin Rombach
    @robrombach
    Apr 19, 2023
    Finally out: our take on text-to-video synthesis with Latent Diffusion Models arxiv.org/abs/2304.08818 An important part is the temporal finetuning of the image decoder - it fixes the flickering artifacts coming with a naive latent approach.
    user avatar
    AK
    @_akhaliq
    Apr 19, 2023
    Align your Latents: High-Resolution Video Synthesis with Latent Diffusion Models turn the publicly available, state-of-the-art text-to-image LDM Stable Diffusion into an efficient and expressive text-to-video model with resolution up to 1280 x 2048 abs: arxiv.org/abs/2304.08818
    Image
    00:00
    arXiv logo
    arxiv.org
    Align your Latents: High-Resolution Video Synthesis with Latent...
    Latent Diffusion Models (LDMs) enable high-quality image synthesis while avoiding excessive compute demands by training a diffusion model in a compressed lower-dimensional latent space. Here, we...
    92K
  • user avatar
    Robin Rombach
    @robrombach
    Mar 3, 2021
    We've just updated our ''Taming Transformers" Github repository github.com/CompVis/taming…: It now includes a colab notebook (colab.research.google.com/github/CompVis…) which can be used to compare different VQGAN configurations for codebook learning -- and OpenAI's DALL-E autoencoder.
    Image
  • user avatar
    Robin Rombach
    @robrombach
    Mar 6, 2024
    Party time! The SD3 paper made it to arxiv: arxiv.org/abs/2403.03206 Key takeaways: - flow matching is very nice. - back to work with @pess_r and a fantastic team ♥️ The paper is full of details on improved flow matching, scaling and engineering. Enjoy!
    Image
    52K
  • user avatar
    Robin Rombach
    @robrombach
    Dec 21, 2024
    beakthrough
    Image
    13K
  • user avatar
    Robin Rombach
    @robrombach
    Feb 6, 2025
    Featuring Flux 1.1 Ultra ♥️
    Image
    Image
    01:16
    user avatar
    Mistral AI
    @MistralAI
    Feb 6, 2025
    Introducing the all new Le Chat: your ultimate AI sidekick for life and work! Now live on web and mobile!
    29K
  • user avatar
    Robin Rombach
    @robrombach
    Aug 1, 2024
    yes.
    user avatar
    Artificial Guy / João Vitor A.
    @artificialguybr
    Aug 1, 2024
    Black Forest Team say they will release a SOTA OSS VIDEO MODEL.
    Image
    32K
  • user avatar
    Robin Rombach
    @robrombach
    Aug 1, 2024
    Replying to @robrombach
    We love Open Source and release FLUX.1 [schnell]: An ultra efficient, 4-step model, coming with an Apache 2.0 license. It closely matches the performance of [dev] and [pro], and is available on Hugging Face here:
    Image
    black-forest-labs/FLUX.1-schnell · Hugging Face
    From huggingface.co
    28K

New to X?

Sign up now to get your own personalized timeline!

Create account

By signing up, you agree to the Terms of Service and Privacy Policy, including Cookie Use.

Terms·Privacy·Cookies·Accessibility·Ads Info·© 2026 X Corp.
Don't miss what's happening
People on X are the first to know.
Log inSign up
Advertisement
Advertisement