AMD launches Helion AI rack for flexible agentic workloads
AMD introduced Helion, an open-architecture AI rack designed for small businesses and enterprises to run agentic AI workloads. The system lets users adjust CPU core density, performance, and power settings to match tasks like personal agents, enterprise apps, or code execution. Dan McNamara, AMD’s SVP of Compute and Enterprise AI, emphasized that agentic AI workloads vary widely in compute…
Key points
- Helion rack lets users customize CPU cores, performance, and power for agentic AI workloads
- Supports up to **32,768 CPU cores** in CPU-only mode or hybrid CPU-GPU setups
- AMD will open Helion specs via the **Open Compute Project (OCP)** with partner support
Helion integrates with existing IT infrastructure, avoiding costly overhauls. AMD offers configurations ranging from a 32,768-core CPU-only rack (using 6th Gen AMD EPYC processors) to CPU-GPU hybrids with AMD Instinct MI350P GPUs. Cooling options include liquid or air cooling, catering to cost, sustainability, or performance priorities. AMD will contribute Helion specs to the Open Compute Project (OCP) and partner with Cisco, Dell, and HPE for complementary technologies. The move aims to lower barriers to AI adoption for businesses of all sizes.
The story so far
8 episodes →- AMD launches Helion AI rack for flexible agentic workloadsthis story
AMD Unveils Helion: A Flexible AI Infrastructure for Diverse Workloads
Small Business Trends · 3 November 2026
Loading the full article…
This text was published by Small Business Trends and written by David Wilson. 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 →- Super Micro’s Q1 revenue jumps 298% to $19.95 billion, beating estimates · 1 src
- SpaceX scales AI cloud to 13GW by 2027, outpacing Google’s 1999-2026 total · 1 src
- Russia orders 10% of data centers reserved for state use amid drone attacks · 1 src
- Mtmd : add Nemotron 3.5 Transcribe support by pskrunner14 · Pull Request #31471 · ggml-org/llama.cpp · 1 src
- OpenAI proposes renaming data centers to ‘AI compute campuses’ amid backlash · 4 src
Comments
via GitHub Discussions