Loading…
Hugging Face
AI platform and community providing machine learning models, datasets, libraries, and tools for building AI applications.
50 postsWebsite
Latest articles
Introducing swift-huggingface: The Complete Swift Client for Hugging Face
Hugging Face released swift-huggingface, a dedicated Swift package offering complete Hub API integration, reliable file downloads, and inference provider access. The library addresses previous limitations in swift-transformers 1.0, where interrupted multi-gigabyte model downloads could not resume and cache structures differed from Python. To resolve cache duplication, swift-huggingface implements a Python-compatible content-addressed storage layout using symlinks and flock file locking. Authentication is standardized through a TokenProvider pattern supporting auto-detection, Keychain integration, static CI/CD tokens, and OAuth 2.0 sign-in with automatic token refresh. The package also provides URLSession-backed snapshot downloads with granular progress tracking and will soon replace the HubApi implementation inside swift-transformers.
MatttDeepMath: A lightweight math reasoning Agent with smolagents
Mathematical reasoning in large language models often suffers from lengthy chain-of-thought traces and frequent arithmetic mistakes. To address these issues, DeepMath pairs a Qwen3-4B Thinking base model with a sandboxed Python execution environment built using the smolagents library and vLLM backend. The system offloads deterministic calculations by emitting concise Python snippets, executing them safely with restricted module imports and no network access, and folding returned values back into the context. Training utilizes Group Relative Policy Optimization on the Tool-Integrated Reasoning subset of OpenMathReasoning with temperature scheduling and length constraints. Across benchmarks including MATH500, AIME, HMMT, and HLE, DeepMath reduces output token lengths by up to 66% while increasing overall problem-solving accuracy.
Daniel Fleischer, Moshe Berchansky, Moshe WasserblatWe Got Claude to Fine-Tune an Open Source LLM
Hugging Face Skills equips coding agents like Claude Code, OpenAI Codex, and Google's Gemini CLI to manage end-to-end language model fine-tuning. Using the hf-llm-trainer skill, an agent validates dataset formats, selects cloud hardware, configures authentication, and submits jobs to Hugging Face Jobs. Supported techniques include Supervised Fine-Tuning (SFT), Direct Preference Optimization (DPO), and Group Relative Policy Optimization (GRPO) for models ranging across various parameter sizes. The integration incorporates Trackio for real-time monitoring and automates LoRA configuration for larger models. Once training completes, the agent pushes artifacts to the Hugging Face Hub and can convert models to GGUF format for local execution.
ben burtenshaw, shaun smithTransformers v5: Simple model definitions powering the AI ecosystem
Transformers v5.0.0rc-0 introduces major architectural updates focused on simplicity, training, inference, and ecosystem interoperability across modern AI workflows. The release adopts a modular modeling approach and centralizes attention implementations into a unified AttentionInterface abstraction to reduce contribution and code review overhead. Support for Flax and TensorFlow is officially sunset in favor of focusing on PyTorch as the primary backend, while tokenization is standardized around the tokenizers library. For execution workloads, v5 adds native continuous batching, paged attention mechanisms, and a dedicated transformers serve OpenAI-compatible serving system. Finally, weight loading is refactored to make low-precision quantization a first-class citizen alongside broad interoperability with formats such as GGUF, MLX, and TorchAO.
Lysandre, Arthur Zucker, Cyril Vallez, Vaibhav SrivastavDiffusers welcomes FLUX-2
Diffusers introduces support for FLUX-2 models through the Flux2Pipeline and Flux2Transformer2DModel classes. Running the 4-bit quantized checkpoint diffusers/FLUX.2-dev-bnb-4bit requires loading Mistral3ForConditionalGeneration as the text encoder and the transformer in bfloat16 precision with CPU offloading enabled. During image generation with a prompt and 50 inference steps, the pipeline encodes text embeddings through the Mistral 3 model before executing diffusion operations. However, executing the pipeline on a GPU with 14.56 GiB capacity triggers a CUDA OutOfMemoryError during 4-bit dequantization inside bitsandbytes matrix multiplication operations. Additionally, the execution logs warn that Flax classes are deprecated in Diffusers and will be removed in version 1.0.0.
YiYi Xu, Daniel Gu, Sayak Paul, Alvaro Somoza, Dhruv Nair, Aritra Roy Gosthipaty, Linoy Tsaban, Apolinário from multimodal AI artOVHcloud on Hugging Face Inference Providers 🔥
OVHcloud is now integrated as a supported Inference Provider on the Hugging Face Hub, expanding serverless inference options across model pages and client SDKs. The integration enables access to open-weight models, including gpt-oss, Qwen3, DeepSeek R1, and Llama, using European infrastructure with pay-per-token pricing starting at €0.04 per million tokens. Developers can connect via Python and JavaScript SDKs using either direct custom API keys or automatic routing through Hugging Face tokens. The service supports structured outputs, function calling, multimodal workflows, and embedding models while providing sub-200ms first-token response times. Calls routed through Hugging Face pass through standard provider pricing without markups, and PRO tier subscribers receive monthly inference credits.
Gilles Closset, Fabien Ric, Elias Tourneux20x Faster TRL Fine-tuning with RapidFire AI
Hugging Face TRL now officially integrates with RapidFire AI to accelerate LLM fine-tuning and post-training experimentation through concurrent configuration evaluation. Instead of executing training configurations sequentially, the platform shards datasets into discrete chunks and cycles multiple model configurations across GPUs at chunk boundaries. The runtime handles automatic checkpointing via a shared-memory-based adapter and model spilling mechanism to maintain consistent training state. Experimenters can monitor live metrics on an MLflow-based dashboard and use Interactive Control Ops to stop underperforming runs or clone and warm-start promising configurations mid-flight. Benchmark results on NVIDIA A100 GPUs demonstrate experimentation speedups ranging from 15x to 20x over sequential runs when training TinyLlama-1.1B and Llama-3.2-1B models.
Kamran Bigdely, Arun Kumar, Quentin GallouédecOpen ASR Leaderboard: Trends and Insights with New Multilingual & Long-Form Tracks
The Open ASR Leaderboard expanded its evaluation framework by introducing dedicated multilingual and long-form transcription tracks across eleven datasets. Benchmarking more than sixty open- and closed-source models from eighteen organizations revealed clear operational trade-offs between accuracy and inference speed. Systems combining Conformer encoders with large language model decoders achieved the lowest word error rates in short-form English speech recognition. In contrast, connectionist temporal classification and token-and-duration transducer decoders yielded tenfold to hundredfold improvements in throughput with only slight accuracy penalties. While closed-source commercial systems maintained performance leads in long-form audio transcription, specialized single-language models consistently outperformed generalized systems supporting hundreds of languages.
Eric Bezzam, Steven Zheng, Eustache Le Bihan, Vaibhav SrivastavIntroducing AnyLanguageModel: One API for Local and Remote LLMs on Apple Platforms
Apple developers frequently face integration friction when trying to support a mix of local and cloud language models across disparate APIs. To resolve this fragmentation, AnyLanguageModel introduces a unified Swift package that acts as a drop-in replacement for Apple's Foundation Models framework. The package standardizes interactions across local backends such as Core ML, MLX, and llama.cpp alongside cloud providers like Anthropic and OpenAI. To avoid dependency bloat from multi-backend support, the library employs Swift 6.1 package traits so projects only import the specific runtime engines they require. Furthermore, the library extends beyond current Foundation Models limitations by adding prompt image support for vision-language models.
MatttApriel-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, merve