Urgent.News

What's breaking now, across thousands of outlets.

Tech

AMD's 192 GB Gorgon Halo prices might leave you petrified

Regular pricing starts at $6,799 - all that memory doesn't come cheap

AMD's 192 GB Gorgon Halo prices might leave you petrified

AMD's Gorgon Halo system-on-chip (SoC) platform has recently arrived, featuring the highest memory capacity available on the market with up to 192 GB of unified memory. This impressive feat makes it possible to run models as large as DeepSeek V4 Flash or Z.AI's GLM-5.3-Flash on a desktop computer, but at a significant cost of $6,799.

For AI enthusiasts and privacy-conscious small businesses, this memory capacity could be invaluable for running larger, more powerful models within the security of their homes or offices. With 4-bit precision, Gorgon Halo can handle around 345 billion parameters on Linux or 320 billion parameters on Windows. The difference in memory allocation lies in how the Linux kernel and AMD's GPU drivers manage the memory partitioning.

The chip, an overclocked version of AMD's Strix Halo APU, boasts 16 Zen 5 cores, a 40-compute-unit integrated GPU, and an XDNA 2-based NPU to power Copilot+ features. While the clock speed increase is modest at about 100 MHz, the faster 8,533 MT/s LPDDR5x memory could provide a slight boost for local AI enthusiasts. However, the current memory shortage is causing prices to soar, making Gorgon Halo-based products significantly more expensive than previous models.

LPDDR5x memory is in short supply, and AMD's embrace of this technology for its own AI-optimized Verano Epycs is only expected to exacerbate the situation. As a result, memory availability and pricing remain a significant challenge for AMD and its competitors in the AI workstation market.

Written by urgent.news from The Register Science's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Also reported by 1 other outlet

Read the original at theregister.com →

More in Tech

AMD Boosting AI/LLM Performance For Radeon iGPUs As Much As 18~23% With Linux 7.4

If you have an AMD Ryzen laptop/desktop and looking to leverage AI/LLM capabilities with the integrated graphics, Linux 7.4 is going to be a real treat especially for lower-end hardware.

  • AMD boosts AI/LLM performance for Radeon iGPUs by 18-23% in Linux 7.4
  • PerfOpt feature bypasses IOMMU for system memory access, reducing overhead
  • Test shows PerfOpt improves local AI/LLM inferencing on AMD Strix Halo desktop

More from Tuesday 29 September →