AMD benchmarks Gorgon Halo AI chips against Intel, ahead of RTX Spark launch
AMD shared benchmarks for its new Gorgon Halo chips, positioning them as a competitor to upcoming RTX Spark devices launching later this week. The company compared its Ryzen AI Max+ Pro 495 to Intel’s Core Ultra X9 388H using ComfyUI, measuring generative AI performance with models like GLM 5.3 Flash and Qwen 3.8 Flash Next. AMD reported throughput advantages ranging from 1.1x to 32.2x, though…
Key points
- AMD benchmarks Ryzen AI Max+ Pro 495 against Intel’s Core Ultra X9 388H, showing 1.1x–32.2x AI throughput gains
- Gorgon Halo supports **192GB** unified memory vs RTX Spark’s **128GB**, but slower token speeds for larger models
- First Gorgon Halo devices (e.g. Minisforum MS-S1 Max-P495) start at **$7,099**, with performance declines at longer context lengths
AMD claims to have shipped ‘over half a million’ AI PCs, though it initially claimed ‘10s of millions.’ The Gorgon Halo range is a refresh of Strix Halo, with minor upgrades like a 100 MHz boost clock increase. First devices, such as the Minisforum MS-S1 Max-P495, start at $7,099. AMD notes performance drops with longer context lengths, which could limit usability for larger models.
AMD attempts to get ahead of expected RTX Spark launch with Gorgon Halo benchmarks
Tom's Hardware · 5 October 2026
Loading the full article…
This text was published by Tom's Hardware and written by Jake Roach. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Hardware & Compute
All →- Open-source voice assistant Speakrail runs locally on an RTX 4090 · 1 src
- ResearchAndMarkets adds GPUaaS forecast: 27.2% CAGR to $118.8B by 2036 · 1 src
- Nvidia partners with Microsoft and Valar to speed up US nuclear builds · 1 src
- Huawei's Atlas 960 Super Pod cuts power use by over 550 kW per machine · 1 src
- Nvidia raises Shield TV Pro price by $100 to $299.99 due to AI-driven component costs · 5 src
Comments
via GitHub Discussions