Meta Platforms runs all Llama inference workloads on Advanced Micro Devices, Inc.’s MI300X, validating its 192GB HBM3 memory and cost-efficiency over Nvidia Corporation. AMD’s data center ...

Private cloud provider Vultr is the first to deploy the AMD Instinct MI325X The move beats the likes of Google and Microsoft ...

The MI300x chip series and ROCm software suite are enhancing AMD’s standing in the AI space, surpassing competitors like Nvidia in some performance metrics. Despite appearing pricey, AMD's rapid ...

The focus on AMD is squarely on upside related to its MI300X accelerator chips, analysts led by Aaron Rakers wrote in an investor note. They see a path towards AMD generating $8B in revenue from ...

It claims that the MI300X GPUs, which are available in systems now, come with better memory and AI inference capabilities than Nvidia’s H100. AMD said its newly launched Instinct MI300X data ...

AMD plans to release a new Instinct data center GPU later this year with significantly greater high-bandwidth memory than its MI300X chip or Nvidia’s H200, enabling servers to handle larger ...

A family of AI chips from AMD. Introduced in 2023, the Instinct MI300X is a GPU chip with 192GB of HBM3 memory. Multiple MI300X chips are used in AMD's Infinity Architecture Platform for high-end ...

Thanks to seamless drop-in compatibility with the AMD Instinct MI300X and integration with AMD ROCm™ open software, the GPU supports key AI and high-performance computing frameworks, simplifying ...