AMD Launches Helios Rack for AI Infrastructure
AMD debuts its co-engineered rack-scale AI solution for data centers at the 2026 Open Compute Project keynote.
Updated on Oct. 22, 2026 in Data Centers
Live Poll
Do you believe the rapid expansion of artificial intelligence infrastructure will improve our society?
AMD has begun shipping its new Helios rack system to customers following a keynote presentation at the 2026 Open Compute Project in San Jose. The 7,000-pound rack represents a collaboration with Meta to address the surging demands of AI workloads.
Why it matters
The company aims to accelerate innovation for joint customers by leveraging an open ecosystem approach for hardware development. This strategy arrives as industry metrics reveal massive growth in agentic traffic and inference-based AI operations.
The Helios system is a 7,000-pound rack co-engineered with Meta that utilizes open rack wide components and specialized power shelves. It is designed to handle an environment where training flops have increased fivefold annually since 2020.
The players
AMD
AMD is a global semiconductor company that develops computer processors and related technologies for data centers and personal computing.
Meta
Meta is a technology conglomerate that develops social media platforms and large-scale AI infrastructure for global user bases.
Open Compute Project
The Open Compute Project is a collaborative community focused on redesigning hardware technology to efficiently support the growing demands on compute infrastructure.
The details
AMD and Meta co-engineered the Helios hardware to manage modern AI scaling, which includes supporting a monthly volume of 40 quadrillion inference tokens. The system is currently entering a production ramp to meet the expanding requirements of data center operators.
Timeline
2020: Training flops growth trajectory began.
2026: AMD delivered a keynote address at the Open Compute Project.
Q4 2026: The production ramp for Helios systems begins.
Q1 2027: The production ramp for Helios systems continues.
2030: Projected target year for data center CPU and AI accelerator TAM milestones.
The Tech Race
The release of the Helios rack reflects the industry's shift toward standardized, high-density AI infrastructure managed through the Open Compute Project. This hardware evolution signals a transition from legacy server architectures to massive, co-engineered systems designed to dominate the projected $1.4 trillion AI accelerator market by 2030.
As data centers adopt high-performance hardware like Helios, users can expect more responsive and capable AI-driven applications. Improved infrastructure efficiency helps companies scale complex services while managing the rising demand for inference-based operations.
The takeaway
The rapid 14x growth in agentic traffic this year highlights a major shift in how organizations prioritize AI processing. Businesses should prepare for higher power and cooling requirements as their local and cloud data center environments integrate heavier rack solutions.
Further reading
Learn more about evolving Data Centers and hardware infrastructure.
Source note: This article includes information reported by DQ.
Live Poll
Do you believe the rapid expansion of artificial intelligence infrastructure will improve our society?
