Compression Enabled MRAM Memory Chiplet Subsystems for LLM Inference Accelerators
By Andy Green (Numen) and Nilesh Shah (Zeropoint Technologies)
This session unveils an AI-specific chiplet alternative to GPU-based inference that can deliver HBM-like bandwidth at 30-50% lower power.
Related Chiplet
- Optical Chiplet
- Reverse Gearbox Chiplet
- FlexGen Multi-Die Smart Network-on-Chip (NoC) IP
- Ncore Multi-Die Interconnect IP
- Integrated voltage regulator (IVR) chiplet
Related Videos
- Chiplet Quilting for the Age of Inference
- Standards for Chiplet Design with 3DIC Packaging (Part 1)
- Standards for Chiplet Design with 3DIC Packaging - Part 2
- Intel Demonstrates First Fully Integrated Optical I/O Chiplet for More Scalable AI
Latest Videos
- The European Chip Ecosystem
- Scaling up AI Compute, Memory and Connectivity
- CURIE: A Cycle-Accurate and Synthesizable UCIe Interface for the Agile Design and Co-Simulation of Chiplets
- Federated Chiplet Design and Simulation
- Custom HBM as a System: Co-designing the Base Die, Die to Die Power, and Package Thermals for AI Accelerators