Similarity Search Engine Accelerate AI Inference
Splitting out the search portion from an AI model means it can be hardware-accelerated to dramatically reduce latency...
Homegrown CNN Accelerator Moves AI to the Edge
Maxim plans to combine its new CNN accelerator with its other IP for a range of AI-enabled application-specific SoCs...
AI Accelerators Promise Power Efficient for Near-Edge Computing
Qualcomm’s Cloud AI 100 silicon will ship next year on three different cards, each…
Co-optimized HW/SW Accelerator for AI Recommendation Models
Hardware-software solution uses sparsity to accelerate AI recommendation models that are typically memory-bound...
France’s Most Promising Startups
France, under La French Tech 120 program, has unveiled the identity of the 123 high-potential startups that will benefit from…
AI Chip Ups the Ante, uses FD-SOI
GreenWaves tweaks architecture, uses FD-SOI to cut power by 5x
AWS Rolls Out Inferentia Chip
Amazon makes Inferentia chip available for customer machine-learning workloads.
AI: There’s More to Chip Performance Than TOPS
What is the best measure of inference performance?
Graphcore’s AI Chips Used in Microsoft’s Azure Cloud
Modest performance improvements on BERT, more dramatic on ResNext.
Nvidia Reveals World’s Smallest AI ‘Supercomputer’
The board delivers 1.4 TOPS/W for drones and small robots.