Franck Abgrall

Sometimes it might be hard to understand the differences between the different web rendering solutions (SSR, CSR, universal rendering, static rendering) Here are animations ✨ explaining in a simplified manner the process of each solution from a user's perspective. [THREAD] 👇

Hugging Face Spaces, the world's largest AI app directory, is now MCP-compatible 🤯 Here turning an entire website into Ghibli in one shot with Claude Code to demonstrate what's now possible: instantly access any A I tool right in your LLM client!

Today is a big day, we're introducing the first version of the HF MCP server 🔥 🧵

🚀 DeepSeek-V3-0324 is out now! 🔹 Major boost in reasoning performance 🔹 Stronger front-end development skills 🔹 Smarter tool-use capabilities ✅ For non-complex reasoning tasks, we recommend using V3 — just turn off “DeepThink” 🔌 API usage remains unchanged 📜 Models are now released under the MIT License, just like DeepSeek-R1! 🔗 Open-source weights: huggingface.co/deepseek-ai/De…

hugging face ecosystem is packed with something for everyone in machine learning (MLE, researcher, builder) 💗 so we made it easier for you to find what you need at /welcome ⤵️

Major milestone in Reinforcement Learning 🚀 In under just 1 hour of training, our AI Agent can now automate tasks requiring complex physics and object interactions. Done for the first time on 100$ robot arms. Watch as it recovers when the cube gets stuck 🤯

Are your ready for a faster Hugging Face? 1. tweak one row in 2TB dataset 2. push takes a coffee sip instead of a lunch break 🤯 ⬇️ join the waitlist today

BOOM! Huge update for AI app developers by @huggingface you can now deploy models directly from Hugging Face with Gradio while choosing your preferred inference provider from @SambaNovaAI, @hyperbolic_labs, @togethercompute, @FireworksAI_HQ, @replicate, @FAL, @nebiusai, @novita_labs Additionally, you can choose to require Space users to log in, ensuring that inference usage is billed to their accounts

Our friends at Hugging Face added video model support to their inference widget. Now you can generate videos right from their website! First up is Wan-AI/Wan2.1-T2V-14B (text to video):

Deepseek R1 just became the most liked model ever on @huggingface just a few weeks after release - with thousands of variants downloaded over 10 million times now!

The Hugging Face Agents Course is now available! Join Alfred on his adventures as he assists Bruce Wayne (you). 😅 hf.co/learn/agents-c…

You can now filter models on Hugging Face that offer inference through Together AI – and run inference on them directly on the model pages!

Announcing HF Spaces' biggest update 🤯 → Ask anything you want to do with AI. With ~400k Spaces, this is the best place to find the AI apps you need ✅

The new HF spaces page is a great way to see what open models are capable of

The burst of DeepSeek V3 has attracted attention from the whole AI community to large-scale MoE models. Concurrently, we have been building Qwen2.5-Max, a large MoE LLM pretrained on massive data and post-trained with curated SFT and RLHF recipes. It achieves competitive performance against the top-tier models, and outcompetes DeepSeek V3 in benchmarks like Arena Hard, LiveBench, LiveCodeBench, GPQA-Diamond. 📖 Blog: qwenlm.github.io/blog/qwen2.5-m… 💬 Qwen Chat: chat.qwenlm.ai (choose Qwen2.5-Max as the model) ⚙️ API: alibabacloud.com/help/en/model-… （check the code snippet in the blog） 💻 HF Demo: huggingface.co/spaces/Qwen/Qw… In the future, we not only continue the scaling in pretraining, but also invest in the scaling in RL. We hope that Qwen is able to explore the unknown in the near future! 🔥 💗 Thank you for your support during the past year. See you next year!

JUST IN: Inference Providers at @huggingface Hub Use @FAL @SambaNovaAI @togethercompute @replicate on @huggingface to infer gigantic models like DeepSeek R1 from model page or through our client SDKs (Python, JS)/HTTP calls

🚀 DeepSeek-R1 is here! ⚡ Performance on par with OpenAI-o1 📖 Fully open-source model & technical report 🏆 MIT licensed: Distill & commercialize freely! 🌐 Website & API are live now! Try DeepThink at chat.deepseek.com today! 🐋 1/n

🔥 Exciting update on @huggingface ! We're making it easier for you to stay on top of what matters most. The notifications page now includes filters, allowing you to quickly sort your notifications by category. Stay organized and never miss an important update again! 👀

in other unrelated news, Organizations can now publish blog Articles on @huggingface huggingface.co/blog/huggingfa…

@not_so_lain @realmrfakename No early access, we just renamed from "run with Docker" to "Run locally" 🙂 What's new is that we now show snippet to run the space locally without Docker ("Clone (git)" tab)

@realmrfakename you might have earlier access than me :v but yeh it has been officially launched today on the hub

new feature on the hub 🔥🚀

@m_chirculescu @not_so_lain It's only available for spaces (non-static ones)

@not_so_lain I can't see it :/ .What exactly does it do? Does it only show code to copy/paste or it does more than that?