Intel discontinues open-source Gaudi driver code

Intel has quietly ended support for its open-source user-space Gaudi driver code. The move affects the SynapseAI project related to the company's AI accelerators. This development was reported by Phoronix, a site focused on Linux hardware.

Intel's decision to discontinue the open-source user-space Gaudi driver code marks a shift in its approach to supporting AI hardware on Linux platforms. The Gaudi series represents Intel's line of AI accelerators, and the driver was part of efforts to enable open-source compatibility for these components. Phoronix, known for its coverage of Linux hardware reviews and benchmarks, highlighted this quiet discontinuation without additional details on the timeline or reasons.

The keywords associated with the report include Linux hardware benchmarks, open-source graphics, and Ubuntu hardware, underscoring the technical context within the Linux ecosystem. This includes references to the Phoronix Test Suite, a tool for performance testing. No specific quotes from Intel were provided in the available information, leaving the implications for users relying on Gaudi for AI workloads unclear.

As a result, developers and users may need to explore alternative drivers or proprietary solutions for Gaudi integration. The event aligns with broader trends in open-source hardware support, though Intel has not issued public statements on the matter.

Связанные статьи

Intel will ship a new graphics processing unit designed for AI inference tasks by the end of this year. The chip uses lower-cost memory and air cooling to undercut rivals Nvidia and AMD.

Сообщено ИИ

The Fedora Council has approved a new plan to create specialized AI-focused desktop images for developers. The effort comes shortly after similar moves by Ubuntu and targets support for major hardware platforms including Intel, AMD, NVIDIA, and ARM.

Этот сайт использует куки

Мы используем куки для анализа, чтобы улучшить наш сайт. Прочитайте нашу политику конфиденциальности для дополнительной информации.
Отклонить