AI COMPUTE•NVIDIA•GPU CLOUD•DATA CENTERS•SEMICONDUCTORS•ENERGY•FUNDING•M&A•AI COMPUTE•NVIDIA•GPU CLOUD•DATA CENTERS•SEMICONDUCTORS•ENERGY•FUNDING•M&A•
GPU·DATA HUB
Newsletter
AI Infrastructure

Meta launches an open-weight AI model that runs on Macs

Meta and Apple deliver open-weight model engineered for on-device Apple silicon inference

By GPU Data Hub Desk · GPU Data Hub·Editorial standards··United States

Meta has introduced an open-weight model named Muse Spark Open in collaboration with Apple. The release is engineered to execute inference natively on Mac computers powered by Apple silicon. The initiative targets broader local execution of capable neural networks directly on consumer hardware.

Original source

This summary was written by the GPU Data Hub desk from reporting published by TechCrunch on 15 Oct 2026, 20:04. Read the full article at the original publisher.

Read Original Source →

By subscribing you consent to receive the daily briefing. Unsubscribe at any time. See our privacy policy.

Companies

MetaApple
LinkedInX

More from GPU Data Hub

Latest →

CERN and Mistral Plan Foundation Models for Scientific Discovery

Particle physics centre CERN has formed a long-term strategic alliance with European artificial intelligence specialist Mistral to design foundation models aimed at scientific discovery. The initiative links CERN's research data and extensive compute capabilities with Mistral's model-building expertise. Both parties have signed a formal statement of intent to advance the collaboration.

AI Infrastructure··HPCwire

Anthropic taps nonprofits METR and SecureBio to help assess risks and safety practices

Artificial intelligence firm Anthropic has enlisted independent non-profit organisations METR and SecureBio to evaluate safety measures and potential hazards surrounding its models. The company is concurrently holding preliminary consultations with additional external entities, including Hugging Face. The initiative expands external auditing around model development practices.

AI Infrastructure··CNBC Technology

Brookhaven Lab: ModSim 2026 Models What Comes Next for AI and HPC

Brookhaven National Laboratory has hosted the final meeting of the ModSim workshop in Seattle, bringing together national laboratories, universities, and commercial partners. The 15-year programme has focused on the simulation, design, and optimisation of advanced scientific computing architectures suited for artificial intelligence and high-performance computing workloads.

AI Infrastructure··HPCwire

Meta and Apple team up to make Muse AI models run better on Macs

Meta has introduced its Muse Spark Open artificial intelligence model, which it designates as the foremost open-weight model produced in the United States. A joint effort with Apple enables the system to run locally on desktop configurations pairing two M5 Ultra Mac Studio machines via the MLX framework. The partnership highlights an expanding effort to make high-end foundation models accessible on advanced client workstations.

AI Infrastructure··The Verge

ORCA Launches PT-3 Photonic Quantum Computer for AI and Optimization

ORCA Computing has announced the PT-3, the latest generation of its photonic quantum computing hardware aimed at commercial-scale artificial intelligence and optimisation problems. The inaugural machine has been deployed in London at the Digital Realty Innovation Lab, where it operates alongside Nvidia accelerated computing infrastructure. The hybrid installation demonstrates how quantum processors can work directly in concert with enterprise data centre accelerators.

AI Infrastructure··HPCwire