Hugging Face stories

Nvidia is pitching Jetson as the key edge platform for open generative AI, pushing speech, vision and robotics models out of the data centre.

AMD shifts its AI pitch from raw silicon to open software and cloud access as it targets developers and a share of looming trillions.

Multiverse unveils HyperNova 60B 2602, a free, compressed GPT-4-class model promising 50% less memory with near-matching tool performance.

Elastic launches compact multilingual embedding models for Elasticsearch, promising faster, cheaper semantic search and stronger RAG performance.

Linux Foundation unites Anthropic, Block and OpenAI in new Agentic AI hub to standardise tools and protocols for autonomous AI agents.

Abu Dhabi's TII unveils Falcon H1R 7B, a compact open AI model it says rivals much larger rivals in reasoning power while using less compute.

Linux Foundation launches Agentic AI Foundation to standardise open protocols for interoperable AI agents with backing from major tech firms.

OutSystems' Agent Workbench has attracted 1,500 developers as enterprises build over 5,500 AI agents, automating workflows for firms like Axos Bank and Thermo Fisher.

OutSystems sees surging enterprise use of its Agent Workbench, with 5,500 AI agents in development and 1,500 certified developers driving automation.

Hugging Face and Google Cloud expand their partnership to enhance access and deployment of open AI models on Google's cloud infrastructure.

Dell launches the PowerEdge XE7740 server, the first to feature Intel's Gaudi 3 AI accelerators, boosting scalable AI performance for enterprises.

JFrog launches AI Catalogue to help firms securely govern, discover, and deploy AI models, integrating open-source and commercial solutions with enhanced compliance.

Enterprise use of generative AI platforms has surged 50%, with shadow AI posing rising data security risks as unsanctioned apps operate beyond IT control.

A critical flaw in mcp-remote lets attackers hijack AI client systems by executing arbitrary OS commands, urging users to update to version 0.1.16 immediately.

NVIDIA and Black Forest Labs have optimised the FLUX.1 Kontext AI image editing model for RTX GPUs, doubling performance and simplifying workflows with natural language prompts.

Tenable plans to acquire Apex Security, enhancing its AI risk control and exposure management platform amid rising cyber threats from AI adoption.

Red Hat launches llm-d, an open source project for scalable generative AI inference, backed by Google Cloud, IBM, NVIDIA and others to support any model, hardware or cloud.

Red Hat unveils AI Inference Server to boost generative AI performance and efficiency across hybrid cloud platforms with broad model and hardware support.

Alibaba launches Wan2.1-VACE, an open-source AI model combining video generation and editing, streamlining content creation for various media platforms.

Alibaba has open-sourced Qwen3, a new large language model series with up to 235B parameters, supporting 119 languages and advanced hybrid reasoning.