Share story
Products

Nvidia outlines server requirements for CUDA on RISC-V

Nvidia outlines server requirements for CUDA on RISC-V Image: Primary
Nvidia is working to extend CUDA support to RISC-V CPUs and has outlined requirements for systems that would run it, according to a Hot Chips report. The requirements include an RVA23 CPU, server platform specifications, ACPI, PCIe coherency and peer-to-peer PCIe communication. Nvidia also said it is partnering with SiFive, which plans to demonstrate a CUDA-running system at Hot Chips. The source says most existing RISC-V hardware is unlikely to meet the requirements.
Sources
Chips and Cheese
In this story
Nvidia
Published by Tech & Business, a media brand covering technology and business. This story was sourced from Chips and Cheese and reviewed by the T&B editorial agent team.
Back to Newswire
Keep reading
Full wire
AI Products
AI Products

SoftBank seeks up to $100 billion from Gulf investors for AI

SoftBank Group Corp. is seeking to raise as much as $100 billion from investors in the Gulf region to help finance an expansion of the company's investments in artificial intelligence, the Financial Times has reported....

Bloomberg Technology
AI Science
AI Science

Nullify preprint reports selective LLM forgetting without weight updates

Researchers report in a preprint that Nullify, a method designed to suppress specific memorized information in large language models, matches or surpasses established forgetting baselines on TOFU and MUSE while preserving model ut...

cs.LG updates on arXiv.org
AI Science
AI Science

Preprint reports schema-free generation of valid enterprise test data

Researchers report in an arXiv preprint that their Generalist Populator agent generated enterprise data with 100% constraint satisfaction and 0.88 average marginal fidelity across ten simulated environments without accessing datab...

cs.AI updates on arXiv.org
AI Science
AI Science

Preprint reports task-completion gains from agent-generated interfaces

Researchers report in an arXiv preprint that training an agent to generate interactive interfaces improved a 4B model's Pass@3 task-completion score from 9.33% to 58.00%. Their GenUI-Harness pairs an agent that retrieves informati...

cs.AI updates on arXiv.org