Gradio

Have you tried Remade LoRAs yet? Squish🥞, Inflate🎈, Pirate🏴☠️, Warrior💪 and many more effects! Check out the Gradio app Remade AI created for Wan2.1 LoRAs on Hugging Face: https://lnkd.in/e-x3BUGc

Diffusion as Shader: a versatile model for video generation tasks. App is Live on Hugging Face: https://lnkd.in/gNW8RR6z 💪 Diffusion-as-Shader gives you a 3D-aware Video Diffusion model which comes with various Video Generation Controls (motion transfer, camera control, object manipulation in 3D and more) Diffusion-as-Shader authors have released a gradio UI for all tasks. Simply run the webuidotpy script to launch the slick app: https://lnkd.in/gbiAvNHT

Diffusion as Shader: a versatile model for video generation tasks. App is Live on Hugging Face: https://lnkd.in/gNW8RR6z 💪 Diffusion-as-Shader gives you a 3D-aware Video Diffusion model which comes with various Video Generation Controls (motion transfer, camera control, object manipulation in 3D and more) Diffusion-as-Shader authors have released a gradio UI for all tasks. Simply run the webuidotpy script to launch the slick app: https://lnkd.in/gbiAvNHT

Have you tried Remade LoRAs yet? Squish🥞, Inflate🎈, Pirate🏴☠️, Warrior💪 and many more effects! Check out the Gradio app Remade AI created for Wan2.1 LoRAs on Hugging Face: https://lnkd.in/e-x3BUGc

🤩 Thera, the first state-of-the-art model that allows super-resolution to arbitrary-scale ! App is live on Hugging Face : https://lnkd.in/g5ci-jYX THERA code: https://lnkd.in/gCBEFrVz

🔗 Gradio is Solving the "But I Can't Recreate Your Results" Problem in AI with "Deep Links"! 🤩 👇 If you've worked with AI tools, you know this scenario: Someone shares an amazing creation, but when you try to recreate it—even with their instructions—your results look different. Parameters are missed, settings are slightly off, and frustration follows. We're excited to announce Gradio Deep Links (coming soon): ✅ Create something in a Gradio app ✅ Generate a URL that captures the exact state—all parameters and outputs ✅ Share the link with anyone ✅ They see precisely what you created without redoing your work This isn't just about convenience—it's about reproducibility, collaboration, and making AI more accessible. Deep Links build on our existing share capabilities, but with a crucial difference: they snapshot your application's state, not just the application itself. ❓ Let us know if you think this solves significant friction in AI collaboration for you?

LBM-relighting by Jasper AI. Fast Image-to-Image Translation App live on 🤗 Hugging Face: https://lnkd.in/gXQZX6Aa

🔗 Gradio is Solving the "But I Can't Recreate Your Results" Problem in AI with "Deep Links"! 🤩 👇 If you've worked with AI tools, you know this scenario: Someone shares an amazing creation, but when you try to recreate it—even with their instructions—your results look different. Parameters are missed, settings are slightly off, and frustration follows. We're excited to announce Gradio Deep Links (coming soon): ✅ Create something in a Gradio app ✅ Generate a URL that captures the exact state—all parameters and outputs ✅ Share the link with anyone ✅ They see precisely what you created without redoing your work This isn't just about convenience—it's about reproducibility, collaboration, and making AI more accessible. Deep Links build on our existing share capabilities, but with a crucial difference: they snapshot your application's state, not just the application itself. ❓ Let us know if you think this solves significant friction in AI collaboration for you?

Excited to share our new research at Jasper! 🚀 LBM: Latent Bridge Matching for Fast Image-to-Image Translation A new, versatile and scalable method that relies on Bridge Matching in a latent space to achieve fast image-to-image translation Try out our Hugging Face space for object relighting! ✨ 🤗 Gradio demo: https://lnkd.in/ebdBeEc4 👉 Paper: https://lnkd.in/eaVb5HGz 💻 Repo: https://lnkd.in/edqkyywr 📄 Project page: https://lnkd.in/e_DMzay2 Onur Tasar Sanjeev S. Benjamin Aubin Damien Henry ClipDrop

💥 Open model drop: Gemma goes multilingual, multimodal, and long-context! What's in for you? Gemma 3 models come in four sizes (1B, 4B, 12B, 27B), have a context window of up to 128k tokens, accept images and text, and support 140+ languages. Available on Hugging Face. I'm particularly excited about the smaller models outperforming the already amazing Gemma 2 models.