We shipped 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

This is wild 🤯 We finally have an open-source music generation model that rocks 🚀 huggingface.co/m-a-p/YuE-s1...

🔦 Lights out! Hugging Face homepage now features Dark Mode ✅

TRELLIS hype me so much more than Sora 😱

🤯 Awesome 3D generation: TRELLIS might be the solution we've been waiting for, nearly ready for production! ⬇️ Plus, there’s a free ZeroGPU demo available for everyone to try out!

Introducing: AI Video Composer - powered by Qwen2.5-Coder-32B + FFMPEG🔥 Drag and drop your assets (images/videos/audios) to create any video you want using natural language! huggingface.co/spaces/huggi...

We need more Hugging Face merch 😝

Hey Skeeters, you can add your Sky handle to your @huggingface.bsky.social profile - here's how

mhhh is Ublock origin blocking too much things on bluesky?

Lightricks just announced LTX Video, a new open-source, community-driven model for video generation, and claims 5 seconds of 24fps video at 768x512 in just 4 seconds on an Nvidia H100! ⚡

If you need to integrate an AI chat @bsky.app with open weight models and 100% privacy, we (@huggingface) can help!

Observers: A Lightweight SDK for AI Observability TLDR; - Track and record interactions with AI models - Store observations in multiple backends @huggingface.bsky.social, @duckdb.org or Argilla - Query and analyse your AI interactions with ease GitHub: github.com/cfahlgren1/o...

💫 HuggingChat new default model: Qwen2.5-72B 👩‍🎤 This is the true Open AI: huggingface.co/chat

Check hertz-dev release (apache-2.0!) by Standard Intelligence. It's quite impressive! below is a live conversation between a human and the model! huggingface.co/si-pbc/hertz...

HuggingChat available models list is 🔥🔥 (hf.co/chat) 1. meta-llama/Meta-Llama-3.1-70B-Instruct 2. CohereForAI/c4ai-command-r-plus-08-2024 3. Qwen/Qwen2.5-72B-Instruct (🐐) 4. nvidia/Llama-3.1-Nemotron-70B-Instruct-HF 5. Qwen/Qwen2.5-Coder-32B-Instruct 6. meta-llama/Llama-3.2-11B-Vision-Instruct

Do you know about the Hugging Face Inference Playground? it's quite cool! hf.co/playground

If AI slows down for pre-trained models and moves towards building agents on existing models, then the open source community will win 100% at doing it.

The open source takeover isn't a question of if, but when - and 'when' is coming sooner than expected: 1. Open models achieving parity with closed ones 2. Diminishing returns on massive pre-training.

🚀 Qwen2.5-Coder release is a huge deal: - Matches GPT-4 coding capabilities - Open source & Apache 2.0 licensed - Supports 40+ programming languages - Available in 6 sizes (0.5B-32B) Exciting times ahead! huggingface.co/collections/...