Hugging Face (Twitter)

RT @Thom_Wolf: wow, total BoM cost $660, folks

open-source community >> closed source hyped robots

Hugging Face (Twitter)

RT @LeRobotHF: Almost 10,000 followers here! Let's build the biggest and most active community of Robotics AI builders thanks to open-source!

Hugging Face (Twitter)

RT @Thom_Wolf: 3 trillions tokens finely distilled from more than a petabyte of PDF files

We’ve just released FinePDF, the latest addition to the Fineweb datasets

Hugging Face (Twitter)

RT @cgeorgiaw: 🚨 Big news in ML for biotech 🚨

Today, we're launching the Antibody Developability Prediction Competition with @Ginkgo + @huggingface!

💧 Hydrophobicity
🎯 Polyreactivity
🧲 Self-association
🔥 Thermostability
🧪 Titer

🏆 Up to $60k in prizes
📅 Submit by Nov 1, 2025

‌Hugging Face (Twitter)

RT @charlesbben: Recently finished writing a new blogpost about @PyTorch compilation in ZeroGPU Spaces.

Worth reading if you're interested in learning about :

- PyTorch ahead-of-time compilation
- ZeroGPU internals

https://huggingface.co/blog/zerogpu-aoti

Hugging Face (Twitter)

RT @HuggingPapers: Here's your recap of the hottest AI papers on @huggingface for September 1-7! This week, we dive into LLM comprehension, hallucination, robotics, and more:

- Drivel-ology: Challenging LLMs with Interpreting Nonsense with Depth
- From Editor to Dense Geometry Estimator
- Open Data Synthesis For Deep Research (mentioning @Google Gemini)
- Towards a Unified View of Large Language Model Post-Training
- ELV-Halluc: Benchmarking Semantic Aggregation Hallucinations in Long Video Understanding
- Why Language Models Hallucinate
- Robix: A Unified Model for Robot Interaction, Reasoning and Planning (outperforming @OpenAI GPT-4o & @Google Gemini 2.5 Pro)
- DeepResearch Arena: The First Exam of LLMs' Research Abilities

Hugging Face (Twitter)

RT @rohanpaul_ai: MASSIVE. THE LARGEST open-sourced pdf data just dropped on @huggingface . Finepdfs

3 trillion tokens across 475 million documents in 1733 languages.

This is the largest publicly available corpus sourced exclusively from PDFs, containing about

The data was sourced from 105 CommonCrawl snapshots, spanning the summer of 2013 to February 2025, as well as refetched from the internet, and processed using 🏭 datatrove, huggingface's large scale data processing library.

This carefully deduplicated and filtered dataset comprises roughly 3.65 terabytes of 3T tokens. For PII and opt-out see Personal and Sensitive Information and opt-out.

The dataset is fully reproducible and released under the ODC-By 1.0 license. You will be able to access the reproduction code, ablation and evaluation setup in this GitHub repository soon 👷.

Compared to HTML datasets, despite being only mildly filtered, it achieves results nearly on par with...

Перейти на оригинальный пост

Hugging Face (Twitter)

RT @iScienceLuvr: If you need to know how much time left you have to submit your paper, you can check "AI Conference Deadlines"

before there used to be a separate website maintained by PapersWithCode, but since PapersWithCode was shut down, it's now available on HuggingFace

Hugging Face (Twitter)

RT @mervenoyann: upgrade your transformers 🔥

it comes with insanely capable models like SAM2, KOSMOS2.5, Florence-2 and more 🤝

I built a notebook you can run with free Colab T4 to walk through the API for new models 🙋🏻‍♀️ fine-tuning will follow-up soon!

Hugging Face (Twitter)

RT @MaziyarPanahi: Introducing MultiCaRe, open-source, multimodal clinical case datasets on @HuggingFace by @OpenMed_AI Community. Public and ready for load_dataset.

Images: 160K+ figures/subimages

Cases: 85K de-identified narratives + demographics

Articles: 85K metadata + abstracts

🧵 (1/7)

Hugging Face (Twitter)

RT @Tim_Dettmers: It feels the coding agent frontier is now open-weights:

GLM 4.5 costs only $3/month and is on par with Sonnet
Kimi K2.1 Turbo is 3x speed, 7x cheaper vs Opus 4.1, but as good

Kimi K2.1 feels clean. The best model for me. GPT-5 is only good for complicated specs -- too slow.

Hugging Face (Twitter)

RT @HuggingPapers: Meta researchers just unveiled Set Block Decoding on Hugging Face.

It's a game-changer for language model inference, delivering 3-5x speedup in token generation with existing models.

No architectural changes needed, matches previous performance.

Hugging Face (Twitter)

RT @tomaarsen: ModernBERT goes MULTILINGUAL!

One of the most requested models I've seen, @jhuclsp has trained state-of-the-art massively multilingual encoders using the ModernBERT architecture: mmBERT.

Stronger than an existing models at their sizes, while also much faster!

Details in 🧵

Hugging Face (Twitter)

RT @adrgrondin: I gave SmolLM3 by @huggingface a voice 🗣️

Here’s a demo of me talking with the model hands-free on iPhone, thanks to built-in voice activity detection

Everything runs fully on-device, powered by Apple MLX

‌Hugging Face (Twitter)

RT @daftengine: aaaaand we're live on @huggingface documentation! Thank you to @lhoestq, @vanstriendaniel and the Hugging Face team for all their help pushing this through and excited for our continued collaboration!
na2.hubs.ly/H010TDt0

#Daft #HuggingFace #Multimodal #OpenSource

Hugging Face (Twitter)

RT @vanstriendaniel: Visual-TableQA: Complex Table Reasoning Benchmark

- 2.5K - tables with 6K QA pairs
- Multi-step reasoning over visual structures
- 92% human validation agreement
- Under $100 generation cost