Quickly skimming today's papers

diffusion language models mean that loras would be new system prompts?

Meet HopeJr, a full humanoid robot lowering the barrier to entry! Capable of walking, manipulating many objects, open-source and costs under $3000 🤯 Designed by Rob Knight and Hugging Face 👇

the conference is getting crazy over it today we're unveiling our 1st robot Hugging Face 🤝 Pollen Robotics a low-cost $250 open-source robot designed as an open-source platform for fun human computer interactions powered by HF Spaces-models-community > discord.com/invite/jsvMRQx…

type shi one wanna see

At World Labs, we built a new Gaussian splatting web renderer with all the bells and whistles we needed to make splats a first-class citizen of the incredible Three.js ecosystem. Today, we're open sourcing Forge under the MIT license.

> The greatest victory is that which requires no battle

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

Today we release ScreenSuite, the most comprehensive evaluation suite for GUI agents (aka Computer Use agents). We packed 13 benchmarks, and 3 different environments, to evaluate the full range of agentic capabilities for vision models. And it turns out, Qwen models are

things happen on 🤗 hugging face first! try this with hf.co/mcp - "can you find me the top 50 trending models, and highlight those with demo spaces". get direct links to try the very latest models...

Awni Hannun Julien Chaumond Agreed: just like MCP is standardizing all RAG/tool connectors, standardizing on template "<tool_call> ... </tool_call>" should be priority for all AI engineers/community. After analyzing top 1k llms for hf.co/spaces/hugging…, we have identified 6 unique tool templates so far.

Autocomplete all the way. Need a linux fork (AI OS) in which LLM is a core kernel, not some adhoc addon. Every input (click, touch, keystroke) as part of LLM context; every input channel (textbox, console input, ...) is automcompleted by default Cursor but OS

Running mlx-lm locally with MCP using Hugging Face's tiny-agents is actually pretty easy and works quite well. A demo running Qwen3 4B with an MCP client for the local file system:

Apple being the producer of the most widely used laptops & smartphones while having shitty AI is what is delaying AGI (costing efficiency & GDP loss at a global scale) We need laptops & smartphones with much better AI integrations

On a daily basis, ChatGPT has been way more accurate & useful than Google searches regarding medical questions