Hugging Face
194 subscribers
1.34K photos
408 videos
2.19K links
Download Telegram
Hugging Face (Twitter)

RT @Xianbao_QIAN: This work from @BytedanceTalk seed team will be transformative and open the era of iterative latent reasoning:

why do model have to only think in human languages?
At least I don't.

The result is also significant:
- 2.6B R4 (4 steps) model achieved comparable performance with qwen3 8B and Gemma 3 12B

The project was supervised by @Yoshua_Bengio @jasoneshraghian How do you feel about this work?
This media is not supported in your browser
VIEW IN TELEGRAM
Hugging Face (Twitter)

RT @allen_ai: Introducing OlmoEarth 🌍, state-of-the-art AI foundation models paired with ready-to-use open infrastructure to turn Earth data into clear, up-to-date insights within hoursβ€”not years.
Hugging Face (Twitter)

RT @ClementDelangue: When you run AI on your device, it is more efficient and less big brother and free!

So it's very cool to see the new llama.cpp UI, a chatgpt-like app that fully runs on your laptop without needing wifi or sending any data external to any API.

It supports:
- 150,000+ GGUF models
- Drop in PDFs, images, or text documents
- Branch and edit conversations anytime
- Parallel chats and image processing
- Math and code rendering
- Constrained generation with JSON schema supported

Well done @ggerganov and team!
Hugging Face (Twitter)

RT @victormustar: New: Qwen-Image-2509-MultipleAngles. Very solid model and probably a lot of creative use cases to find with it.

⬇️ Free demo available on Hugging Face
Hugging Face (Twitter)

RT @alvarobartt: One of the best parts of working at @huggingface is seeing how much the community drives open-source forward. On Text Embeddings Inference (TEI), we're just one PR away from 300 contributions thanks to all the people who've helped make embedding inference and serving better, faster, and safer.

Nice reminder that open-source only works because people care enough to build together.
This media is not supported in your browser
VIEW IN TELEGRAM
Hugging Face (Twitter)

RT @multimodalart: Qwen Image Multiple Angles LoRA is an exquisitely trained LoRA! πŸ“Λšβ‚Šβ€§κ’°αƒ

Keep character and scenes consistent, and flies the camera around! Open source got there! One of the best LoRAs I've come across lately πŸ™Œ
This media is not supported in your browser
VIEW IN TELEGRAM
Hugging Face (Twitter)

RT @chichengcc: Can we collect robot data without any robots?

Introducing Universal Manipulation Interface (UMI)

An open-source $400 system from @Stanford designed to democratize robot data collection

0 teleop -> autonomously wash dishes (precise), toss (dynamic), and fold clothes (bimanual)
This media is not supported in your browser
VIEW IN TELEGRAM
Hugging Face (Twitter)

RT @linoy_tsaban: adding camera control to the list of things Qwen Image Edit is great at
+ with a specialized multi-angle LoRA it's even better✨
> rotate the camera
> tilt between bird’s-eye and worm’s-eye views
> adjust lens (wide, close-up)

of course we built a demo for it πŸ€πŸ“Ή
This media is not supported in your browser
VIEW IN TELEGRAM
Hugging Face (Twitter)

RT @andimarafioti: Mixing vision and robotics is incredibly hard, but when it finally works it feels like magic.
Hugging Face (Twitter)

RT @lhoestq: Just released today ! Full support for dataset streaming in large scale distributed training

With datasets==4.4.1 and huggingface_hub==1.1.0, better than 4.4 and 1.0

These include all the features explained in the blog post https://huggingface.co/blog/streaming-datasets

Advanced examples in replies:
Hugging Face (Twitter)

RT @Kimi_Moonshot: πŸš€ Hello, Kimi K2 Thinking!
The Open-Source Thinking Agent Model is here.

πŸ”Ή SOTA on HLE (44.9%) and BrowseComp (60.2%)
πŸ”Ή Executes up to 200 – 300 sequential tool calls without human interference
πŸ”Ή Excels in reasoning, agentic search, and coding
πŸ”Ή 256K context window

Built as a thinking agent, K2 Thinking marks our latest efforts in test-time scaling β€” scaling both thinking tokens and tool-calling turns.

K2 Thinking is now live on kimi.com in chat mode, with full agentic mode coming soon. It is also accessible via API.

πŸ”Œ API is live: platform.moonshot.ai
πŸ”— Tech blog: https://moonshotai.github.io/Kimi-K2/thinking.html
πŸ”— Weights & code: huggingface.co/moonshotai
Hugging Face (Twitter)

RT @hanouticelina: We're definitely not slowing down - it's only been 2 weeks since v1.0, but πš‘πšžπšπšπš’πš—πšπšπšŠπšŒπšŽ_πš‘πšžπš‹ v1.1 is already here packed with a lot of nice features! πŸ”₯
This update brings a faster and cleaner download experience and new powerful features to the Hugging Face CLI!

$ πš™πš’πš™ πš’πš—πšœπšπšŠπš•πš• πš‘πšžπšπšπš’πš—πšπšπšŠπšŒπšŽ_πš‘πšžπš‹ --πšžπš™πšπš›πšŠπšπšŽ

🧡See what's new in the thread below πŸ‘‡
This media is not supported in your browser
VIEW IN TELEGRAM
Hugging Face (Twitter)

RT @Wauplin: Using the hf CLI just got insanely easy:

uvx hf auth login
uvx hf download
uvx hf ...

Always runs the latest version in an isolated environment. No install or updates required thanks to @astral_sh 's awesome tooling!
This media is not supported in your browser
VIEW IN TELEGRAM
Hugging Face (Twitter)

RT @jadechoghari: We're introducing EnvHub to LeRobot!

Upload simulation environments to the @hugginigface hub, and load them in lerobot, with one line of code !

env = lerobot.make_env("username/my-env", trust_remote_code=True)

Back in 2017, @OpenAI called on the community to build Gym environments.
Today, we're doing the same for robotics.

We're calling researchers, students, hobbyists, labs, and startups to publish complex envs, tasks, and benchmarks on the Hub, so we see how robot policies perform. (in Isaaac, Mujoco, Genesis, etc..)
Fill out and submit the form linked in the comments to join the effort!
Hugging Face (Twitter)

RT @lunarflu1: 2025 update: We now have 12 different free courses available on @huggingface !

Check it out here:
hf.co/learn
Hugging Face (Twitter)

RT @abidlabs: I'm working on getting more developers to build software with open-weight models and open-source tooling. Not always easy to change entrenched software developer behavior but if we are able to, a lot of good will come out of it. Looking for help, DM me if you're interested!
Hugging Face (Twitter)

RT @ben_burtenshaw: you can now push, run, and pull agentic environments from the hugging face hub as spaces.

the workflow looks like this:
- do `openenv init` to start a new environment from a template
- build/port your rl environment
- use `openenv push` to push the environment to hub.
- from there, you can either call the api directly
- or, just pull it as a locally running docker container. mmmm

can't wait to see rl environments get super open