Hugging Face in the news

AI reasoning models consume significantly more energy—up to 30 times more on average—due to their step-by-step text generation process.

SciBERT remains a leading scientific NLP model with high adoption, strong benchmark performance, and significant impact in the growing healthcare NLP market.

South Korea's National AI initiative is releasing open-source foundation models on Hugging Face for global testing and competition among top domestic AI firms.

AWS demonstrates hybrid edge-cloud AI agents controlling robots with real-time physical interaction and cloud-based reasoning, enabling continuous learning and scalable deployment.

Hugging Face is growing its AI platform with millions of users and expanding into open-source robotics, integrating machine learning models for robot control and offering hardware like the Reachy Mini.

UMA, a Paris startup founded by ex-Tesla, DeepMind, and Hugging Face experts, is developing industrial and humanoid robots to assist various sectors and reduce import dependency.

Hugging Face provides advanced NLP tools with pre-trained models for tasks like text generation, sentiment analysis, and classification, demonstrated through a Python tutorial.

Hugging Face's AnyLanguageModel library enables seamless switching between local and cloud AI models on Apple devices, simplifying development and enhancing performance.

Hugging Face CEO warns of an impending 'LLM bubble' burst but sees AI's future thriving with specialized models and sustainable growth.

Hugging Face and Google Cloud partnered to improve open-source AI model access, performance, TPU support, and security for developers at scale.

Hugging Face and Google Cloud partner to speed up model downloads, add TPU support, and enhance security for AI developers.

Google Cloud enhances its integration with Hugging Face by improving model loading speeds, adding TPU support, and strengthening security for AI models on Vertex AI.

Hugging Face released a free, interactive robotics course teaching AI and machine learning applications in robotics using their open-source tools.

Meta and Hugging Face introduced OpenEnv, an open-source platform for building secure, standardized environments that safely constrain AI agents' tool access.

Hugging Face launched huggingface_hub v1.0, a major infrastructure upgrade enhancing file transfer, CLI, and enterprise MLOps capabilities, positioning it as a central open machine learning platform.

Hugging Face is building LeRobot, an open-source robotics platform with affordable hardware and software to grow a large community enabling accessible, safe, and innovative physical AI applications.

Epita collaborates with Hugging Face to enhance AI education by integrating open-source tools, resources, and practical learning for students and faculty.

Hugging Face launched HuggingChat Omni, an AI router that automatically selects the best open source model for each user prompt from over 100 models.

Hugging Face offers a vast open-source AI model platform emphasizing community-driven discovery, open science, and the evolving synergy between AI software and hardware development.

Enterprises can cut AI costs and energy use by right-sizing models, optimizing hardware, adopting efficiency defaults, incentivizing energy transparency, and rethinking the need for massive compute clusters.