Loading…
Hugging Face
AI platform and community providing machine learning models, datasets, libraries, and tools for building AI applications.
61 postsWebsite
Latest articles
Apriel-H1: The Surprising Key to Distilling Efficient Reasoning Models
Retrofitting efficiency into existing large language models often fails when naive distillation disrupts complex multi-step reasoning mechanisms. The Apriel-H1 project addresses this challenge by converting a 15B full-attention reasoning model into a Mamba hybrid using targeted distillation on high-quality SFT reasoning traces rather than diffuse pretraining data. Built on the open-source Fast-LLM framework, the process applies reverse KL divergence alongside a staged layer replacement strategy combining Leave-One-Out analysis and dynamic loss evaluations. Across fifty total blocks, replacing thirty attention layers with Mamba mixers yielded the flagship Apriel-H1-15b-Thinker-SFT checkpoint. This configuration achieved a 2.1x throughput increase with minimal benchmark degradation after 76.8B training tokens.
Torsten Scholak, Oleksiy Ostapenko, Raymond Li, Luke Kumar, Joel Lamy-PoirierEasily Build and Share ROCm Kernels with Hugging Face
Compiling custom deep learning GPU kernels across architectures often leads to complex build flag management, compiler errors, and ABI issues. Hugging Face's kernels library and kernel-builder streamline this workflow by supporting backends such as ROCm, CUDA, Metal, and XPU. Using the RadeonFlow FP8 block-wise GEMM kernel optimized for the AMD Instinct MI300X GPU as an example, developers organize files into specific headers and HIP implementations. The build is configured via a build.toml manifest and flake.nix to ensure reproducible compilation across supported PyTorch and ROCm versions. Artifacts can then be pushed to the Hugging Face Hub, allowing users to load and execute custom operators directly via get_kernel without traditional installation steps.
Abdennacer Badaoui, Daniel Huang, colorswind, Zesen LiuJoin the AMD Open Robotics Hackathon
AMD, Hugging Face, and Data Monsters announced the AMD Open Robotics Hackathon, an in-person competition scheduled for Tokyo and Paris in December 2025. Participating teams of up to four adults will compete over three days across two core missions. The first mission covers an instructor-led session on configuring the LeRobot development environment using AMD AI solutions. During the second mission, teams spend two days developing a freestyle project to address a real-world problem using LeRobot. Participants receive access to SO-101 robotics kits, AMD Ryzen AI laptops, and AMD Instinct MI300X GPUs via AMD Developer Cloud, with prizes awarded to the top seven teams in each city based on creativity, difficulty, ease-of-use, and practicality.
Eric Ma, Guruprasad MPBuilding for an Open Future - our new partnership with Google Cloud
Hugging Face and Google Cloud announced an expanded strategic partnership designed to streamline the deployment and management of open models on Google Cloud infrastructure. Prompted by a tenfold increase in Hugging Face usage on Google Cloud over three years, the collaboration introduces a joint CDN Gateway using Hugging Face Xet technology to cache models and datasets directly on Google Cloud. This caching mechanism aims to shorten download times, strengthen model supply chain resilience, and accelerate time-to-first-token across Vertex AI, Google Kubernetes Engine, Cloud Run, and Compute Engine virtual machines. Hugging Face plans to integrate native library support for Google Cloud TPUs, lower instance prices on Inference Endpoints, and enhance Hub security scanning via VirusTotal, Google Threat Intelligence, and Mandiant.
Jeff Boudier, Simon PagezyAligning to What? Rethinking Agent Generalization in MiniMax M2
MiniMax M2 addresses the performance gap between agent benchmark scores and real-world deployment across diverse scaffolding frameworks and environments. While initial alignment efforts relied on scaling tool variety, models continued to fail when subjected to slight variations in prompts, scaffolding, and tool responses. To resolve this instability, the team introduced interleaved thinking, which enables models to execute internal reasoning at arbitrary steps throughout long-horizon tasks to adapt dynamically to external tool outputs. Additionally, developers constructed a data pipeline designed for full-trajectory generalization by introducing perturbations across system prompts, user objectives, environments, and tool feedback. In internal evaluations, MiniMax M2 maintained robust tool-calling and instruction-following performance when tested against unfamiliar, cold-start agent scaffolding frameworks.
MiniMaxOn the Shifting Global Compute Landscape
United States export controls on advanced artificial intelligence hardware have catalyzed rapid expansion across China's domestic silicon and open-weight model ecosystem. Faced with restricted access to high-end NVIDIA GPUs, Chinese developers accelerated deployments on domestic accelerators, including Huawei Ascend, Cambricon, and Baidu Kunlun. Hardware scarcity spurred architectural and algorithmic innovations in compute efficiency, notably DeepSeek's Multi-head Latent Attention and Group Relative Policy Optimization, alongside substantial post-training cost reductions. Organizations such as Baidu and Ant Group now train foundation models directly on domestic hardware, fostering non-CUDA software stacks and lowering inference costs globally. Consequently, the global artificial intelligence infrastructure is shifting from an exclusively American-focused paradigm toward a dual-ecosystem landscape powered by domestic chips and open-weight architectures.
Tiezhen WANG, Irene SolaimanBuilding a Healthcare Robot from Simulation to Deployment with NVIDIA Isaac
Simulation in healthcare robotics has historically been slow, siloed, and difficult to translate into real-world physical systems. The NVIDIA Isaac for Healthcare v0.4 release introduces an end-to-end SO-ARM starter workflow to streamline simulation, training, and deployment for embodied surgical assistants. The architecture relies on a Sim2Real mixed training approach that fine-tunes the GR00T N1.5 foundation model on combined real-world and synthetic datasets using LeRobot. Synthetic data generated in simulation constitutes over 93 percent of the policy training data to bridge physical data collection constraints. Real-time inference on the SO-ARM101 precision manipulator utilizes RTI DDS communication, dual-camera vision, and TensorRT model conversion.
Steven Palma, Andres Diaz-PintoHow to Build a Healthcare Robot from Simulation to Deployment with NVIDIA Isaac for Healthcare
Simulation in healthcare robotics has historically been slow, siloed, and difficult to translate to physical hardware. NVIDIA introduced the Isaac for Healthcare v0.4 developer framework to streamline this lifecycle through an end-to-end SO-ARM starter workflow. Developers collect mixed real-world and synthetic datasets using LeRobot and SO-101 teleoperation, with synthetic generation accounting for over 93 percent of training data. The pipeline post-trains GR00T N1.5 models using dual-camera vision inputs, evaluates policies in Isaac Lab, and deploys them to physical hardware with RTI DDS communication. This integrated approach allows surgical assistant robots to interpret natural language instructions and execute physical manipulation tasks reliably.
AsawareeGranite 4.0 Nano: Just how small can you go?
IBM has released Granite 4.0 Nano, a collection of sub-billion and ~1.5B parameter language models tailored for edge and on-device deployments. Trained on more than 15 trillion tokens under an Apache 2.0 license, the suite includes four instruct models alongside their base counterparts. The lineup introduces hybrid-SSM architectures—specifically the ~1.5B parameter Granite 4.0 H 1B and the ~350M parameter Granite 4.0 H 350M—as well as standard transformer variants suited for runtimes like llama.cpp. According to benchmark results across general knowledge, math, code, and safety domains, these compact models demonstrated strong capability improvements relative to their parameter footprint. Furthermore, evaluation on IFEval and Berkeley's Function Calling Leaderboard v3 indicated competitive performance on instruction following and tool calling tasks.
Kate Soule, Rameswar PandaVoice Cloning with Consent
Realistic voice generation presents significant deepfake risks alongside legitimate communication benefits. To prevent unauthorized voice duplication, a voice consent gate integrates verification directly into the voice cloning workflow. The system relies on a language model to generate novel sentence pairs containing explicit consent and phonetically diverse phrases on random everyday topics. Users must read these generated lines aloud through a direct microphone stream rather than uploading pre-recorded files, and an automatic speech recognition system verifies the utterance against the generated text. Once confirmed, the consenting audio snippet is passed directly to the voice-cloning text-to-speech model as the reference sample for synthesis.
Margaret Mitchell, Lucie-Aimée KaffeeStreaming datasets: 100x More Efficient
Training machine learning models on multi-terabyte datasets often requires long download times or suffers from network request storms when multiple DataLoader workers initialize independently. Hugging Face revamped the backend of its datasets library to make dataset streaming viable for large-scale distributed training without modifying the standard API. Startup latency was reduced by caching resolved data file lists across workers and bundling API calls, cutting initial request volume up to one hundred times. The team also implemented background prefetching for Parquet files and added configurable buffering parameters to keep GPUs fully saturated during training runs. Benchmark tests demonstrated up to twice the sample processing throughput and zero worker crashes across 256 concurrent workers on 64 H100 GPUs.
Andres Marafioti, Quentin Lhoest, ben burtenshaw, Pedro Cuenca, mervehuggingface_hub v1.0: Five Years of Building the Foundation of Open Machine Learning
The Hugging Face Hub team released huggingface_hub v1.0, marking the library's transition into a mature foundation for open machine learning after five years of development. The major update migrates the underlying network backend to httpx and replaces the legacy hf_transfer tool with the hf_xet binary package to enable chunk-level file deduplication across repositories. In addition, the release introduces a redesigned Typer-based CLI using the streamlined hf command alongside explicit authentication functions and expanded inference primitives. Legacy components, including the Git-based Repository class and the older InferenceApi interface, have been removed to establish a cleaner and more maintainable architecture. Most dependent machine learning libraries remain compatible across versions, though transformers requires v0.x for version 4 and v1.x for its upcoming version 5 release.
Lucain Pouget, Célina Hanouti, Lysandre, Julien ChaumondBuilding the Open Agent Ecosystem Together: Introducing OpenEnv
Meta and Hugging Face have partnered to launch OpenEnv Hub, an open platform and specification designed for agentic AI environments. Large language models require external tools and APIs to complete tasks, but exposing broad toolsets directly introduces significant safety and semantic challenges. In response, OpenEnv establishes sandboxed environments providing defined APIs, authenticated tool access, and isolation for reinforcement learning post-training and deployment pipelines. The initial release includes the OpenEnv 0.1 RFC series covering component architecture, packaging, communication, and Model Context Protocol tool encapsulation. Creators can run local Docker-based environments using standard step, reset, and close interfaces while integrating with frameworks like TorchForge, TRL, verl, and SkyRL.
Joseph Spisak, Davide Testuggine, Zach Wentz, Pierre Andrews, Sanyam Bhutani, Hamid Shojanazeri, Pankit Thapar, Emre Guven, Lewis Tunstall, Vaibhav SrivastavHugging Face and VirusTotal collaborate to strengthen AI security
Hugging Face announced a collaboration with VirusTotal to enhance the security of files shared across the Hugging Face Hub. Hosting more than 2.2 million public model and dataset repositories, the platform faces potential threats from malicious payloads, compromised files, and dependencies that execute unsafe code. Starting immediately, public repositories on the Hub undergo continuous scanning powered by VirusTotal threat intelligence. When users view repository or file pages, the Hub checks file hashes against VirusTotal without transferring raw file contents, preserving user privacy. Retrieved metadata displays detection counts, clean or malicious status, and known-bad relationships, enabling organizations to evaluate risks prior to downloading assets or integrating them into deployment pipelines.
Adrien Carreira, Bernardo QuinteroSentence Transformers is joining Hugging Face!
Sentence Transformers is officially transitioning its stewardship from the Ubiquitous Knowledge Processing Lab at TU Darmstadt to Hugging Face. The open-source embedding library was introduced in 2019 by Nils Reimers using a Siamese network architecture to overcome standard BERT limitations on sentence-level semantics. Tom Aarsen of Hugging Face has maintained the project since late 2023 and will continue leading its development. Under Hugging Face, the project retains its Apache 2.0 license and leverages infrastructure such as continuous integration and automated testing. Sentence Transformers currently serves over one million monthly unique users, with more than 16,000 community models publicly hosted on the Hugging Face Hub.
Tom AarsenUnlock the power of images with AI Sheets
Hugging Face released an update to AI Sheets, an open-source tool for transforming and enriching datasets without code. The update adds vision support powered by Inference Providers, enabling users to upload images, extract structured data, generate visuals, and perform image edits in a spreadsheet interface. Each column can execute AI actions configured with custom prompts and selected models. For text extraction and reasoning tasks, AI Sheets supports vision language models including Qwen/Qwen2.5-VL-7B-Instruct and Qwen/Qwen3-VL-235B-A22B-Reasoning, while image transformations use models like Qwen-Image-Edit. Users can iterate on prompts, provide feedback using thumbs-up ratings to create few-shot examples, and export completed datasets to the Hugging Face Hub, CSV, or Parquet.
Ame Vi, Daniel Vila, Francisco Aranda, Damián Pumar, Leandro von Werra, Thomas WolfAI for Food Allergies
Global food allergies affect an estimated 220 million individuals, yet computational progress in biomedical discovery remains constrained by fragmented and inaccessible scientific data. The AI for Food Allergies initiative addresses this challenge by establishing an open research community and releasing the curated Awesome Food Allergy Datasets collection across multiple biological layers. Computational pipelines leverage deep learning models, such as AllergenAI and NetAllergen-1.0, which incorporate sequence motifs and computationally predicted MHC class II presentation propensities to evaluate allergenicity. Additionally, molecular property prediction benchmarks like QM9 provide high-accuracy quantum-mechanical properties for approximately 134,000 molecules, supporting generative and virtual screening workflows targeting IgE–FcεRI binding. These combined efforts systematically structure molecular, clinical, and chemogenomic resources to accelerate allergy diagnostics and therapeutic protein engineering.
Ludovico Comito, Antonis Vozikis, Vaibhav Pandey, Kisejjere RashidGoogle Cloud C4 Brings a 70% TCO improvement on GPT OSS with Intel and Hugging Face
Intel and Hugging Face benchmarked OpenAI GPT OSS text generation to evaluate performance on Google Cloud C4 virtual machines powered by Intel Xeon 6 processors against previous-generation C3 instances. The benchmark evaluated the unsloth/gpt-oss-120b-BF16 Mixture of Experts model across batch sizes from 1 to 64 using bfloat16 precision, static KV cache, and SDPA attention. To eliminate redundant computation, Hugging Face merged an optimization directing each expert sub-network to process only its routed tokens rather than all tokens. Testing demonstrated that C4 instances deliver 1.4x to 1.7x higher throughput per vCPU compared to C3 instances. At a batch size of 64, C4 achieved a 1.7x Total Cost of Ownership improvement alongside lower hourly pricing.
Jiqing.Feng, Matrix Yao, Ke Ding, Ilyas MoutawwakilGet your VLM running in 3 simple steps on Intel CPUs
Deploying Vision Language Models locally presents computational challenges despite providing data privacy and independence from external servers. Optimum Intel and OpenVINO address this by optimizing lightweight models like SmolVLM2-256M for low-resource hardware, including Intel CPUs and GPUs. The process involves exporting the model to the OpenVINO Intermediate Representation format, applying 8-bit weight-only or static quantization, and executing inference with the transformed model. In single-image CPU benchmarks on an Intel Core Ultra 7 265K processor, OpenVINO with 8-bit weight-only quantization reduced time to first token from 5.150 seconds under PyTorch down to 0.247 seconds. The optimized configuration also raised decoding throughput from 0.722 to 63.928 tokens per second.
Ezequiel Lanza, Helena, Nikita, Ella Charlaix, Ilyas MoutawwakilNemotron-Personas-India: Synthesized Data for Sovereign AI
Most open artificial intelligence datasets reflect Western norms and English-only contexts, limiting model effectiveness across India's multilingual and multi-script landscape. To address this shortage of culturally grounded training data, NVIDIA released Nemotron-Personas-India under a CC BY 4.0 license. The synthetic dataset contains 21 million personas across 3 million records spanning 7.7 billion tokens in English, Devanagari Hindi, and Latin-script Hindi. The generation pipeline utilized NeMo Data Designer, an Apache-2.0 licensed Probabilistic Graphical Model for demographic grounding against 2011 Census distributions, and GPT-OSS-120B for narrative text. All records are entirely synthetic to eliminate re-identification risks while enabling fine-tuning for region-aware AI agents and domain-specific copilots.
Kiran Praveen, Utkarsh Vaidya, Evan A, Lipika Ramaswamy, Dhruv Nathawani, Dane Corneil, Yev Meyer