Hugging Face

In case you didn't already know the awesome FLUX model from Black Forest Labs is supported in 🧨 Diffusers. The model has about 12B parameters along with two text encoders. So, it can be burdensome to run. So, we have put together a little gist guiding you on how to run Flux with limited resources. What a model! Check it out here: https://lnkd.in/gr3R9Eza

SF3D is super-fast image to 3D. Generates 3D mesh in 0.5 second🤯 Attached video is not sped up! Demo + Key Details 👇 SF3D: Stable Fast 3D Mesh Reconstruction with UV-unwrapping and Illumination Disentanglement - SF3D is from Stability AI and is based on their previous work TripoSR. - 🤩 Bonus: You can also select an HDR environment to light up your 3D model. - Great work by Mark Boss on the model development and release!🙌 🥳 You can launch a gradio demo in just 3 steps from the SF3D repo: https://lnkd.in/gvpevyez 🤗 OR access the demo on Hugging Face Spaces: https://lnkd.in/gdUFdU5d

Absolutely wild! 🤯 Google DeepMind Gemma 2B  outperforms OpenAI GPT-3.5 on LMSYS Chatbot arena with a score of 1130! 20 months ago, "ChatGPT is a revolution, the most powerful model ever made," and today, you can run a model more preferred than this literally on a toaster!🍞 🚀 Gemma 2B It also ranks higher than: > Microsoft Phi-3 Medium (14B version) > Mistral AI 8x7B Instruct > Mistral AI 7B fine-tunes > Meta Llama 2 70B Test it on Hugging Face: https://lnkd.in/dbn4ZGjg Leaderboard: https://lnkd.in/dA-2CiEi

🚀 Segment Anything Model 2 (SAM 2) is a foundation model from AI at Meta , designed to address promptable visual segmentation in both images and videos. Deploy with Gradio👇 SAM 2 provides strong performance across a wide range of tasks and visual domains. Here is how you can deploy the model with Gradio: ✅ Google Colab: https://lnkd.in/gFjEyNeb ✅ Hugging Face Spaces demo by Piotr Skalski: https://lnkd.in/gwjkC8cs Have fun!!

Trying to catch Andre De Grasse, the reigning 200m Olympic champ from Canada, using AI at Meta's SAM2! How many of you are working on building a demo for video segmentation for SAM2? Let us know how we can help😊

🤗 🔥 Since joining Hugging Face, we've worked hard to ship 2.0. Data quality is what makes or breaks AI and Argilla 2.0 is the data-centric tool for AI makers ✨ What's the most exciting? 🐍 A new shiny Python SDK: unified API for datasets, workspaces, and user management (plus new docs 📚) 🚀 Launching Argilla on Spaces just got 10x faster: run data annotation projects with your team or the entire community in just two clicks! 🥞 Automatic task distribution: configure the number of responses per data point to complete high quality datasets in record time! 🤔 Never heard of Argilla before? We made this blog post for you: https://lnkd.in/dGJJefHU

🔥 AuraSR Upscaler version2 from fal ai works for non-generated images as well & produces more natural-looking results! Learn more👇 Main features of 🔎AuraSR upscaler v2 : - Works with all kinds of real-world photos, even bit blurry or compressed🤠 - Instead of looking at whole pictures during training, it now looks at smaller pieces. - Blends these smaller pieces smoothly so you can't tell seams in between🔎 - Don't add fake details while still improving the image quality - Coupled with @Gradio you can build a nice upscaling webUI for your image workflows in very few lines of code! Multiple ways to try out AuraSR v2 with Gradio: ✅ Google Colab: https://lnkd.in/gDeHT8AR 🤗 Try Hugging Face Demo by @gokay fem Gökay Aydoğan: https://lnkd.in/gpnJmxNG Model on Hugging Face hub with instructions on using it: https://lnkd.in/gY9DVybq

Introducing Artist👩🎨🎨 A training-free text-driven image stylization SD model 🤯 No training, No ControNets, No LoRAs > Given an image and a prompt describing the desired style, Artist produces stylized image > OG Image details are preserved > Just a pretrained SD is needed There are multiple ways to try out Artist👩🎨 with Gradio: 🔥 Build gradio UI for Artist locally on your machine: https://lnkd.in/gE5SzjPF ✅ Google Colab : https://lnkd.in/gjBsY2De 🤗 Try Hugging Face Space by Sylvain F. https://lnkd.in/g2zi7WyH

With larger and larger diffusion transformers coming up, it's becoming increasingly important to have some good quantization tools for them. We present our findings from a series of experiments on quantizing different diffusion pipelines based on diffusion transformers. We demonstrate excellent memory savings with a bit of sacrifice on inference latency which is expected to improve in the coming days. Diffusers 🤝 Quanto ❤️ This was a juicy collaboration between David Corvoysier and myself. Check out the post to learn all about it https://lnkd.in/gttQFu3z

Serverless Inference with Hugging Face and NVIDIA NIM! Excited to announce the latest update on our collaboration between Hugging Face and NVIDIA! Inference as a Service is a new feature for Hugging Face Enterprise Organizations to easily build and prototype AI applications using the latest open LLMs with Serverless Compute. 👀 TL;DR: 🚀 Introducing Hugging Face Inference-as-a-Service powered by NVIDIA NIM 💼 Available for Enterprise Hub organizations 🖥️ Powered by NVIDIA DGX Cloud and NVIDIA NIMs 🤖 Starting with 7 open LLMs, including Llama 3.1 70B & Mixtral 8x22B 💰 Pay-as-you-go (pay only for request time) with $0.0023/second/gpu 🔌 Supports OpenAI API standards for easy adoption Learn more here: https://lnkd.in/dUQNQ4gS