Cassia Proposes ‘Better Math’ for AI Efficiency
By Sally Ward-Foxton, EETimes | October 1, 2025
Startup Cassia wants to ‘do math better’ by approximating mathematical functions and implementing these approximations in smaller, more power efficient hardware IP for AI acceleration.
The idea began as a spare time project for Cassia founder and CEO James Tandon after he came across scientific papers from another group that discussed the accuracy loss associated with approximating certain math functions rather than quantizing (reducing numerical precision, which reduces prediction accuracy).
To read the full article, click here
Related Semiconductor IP
- Dataflow AI Processor IP
- Powerful AI processor
- AI Processor Accelerator
- High-performance AI dataflow processor with scalable vector compute capabilities
- AI DSA Processor - 9-Stage Pipeline, Dual-issue
Related News
- PGC Strengthens Cloud and AI ASIC Acceleration with Synopsys’ Next-Generation Interface and Memory IP on Advanced Nodes
- Ceva Delivers Real-Time AI Acceleration on NXP’s Processors for Software-Defined Vehicles
- Omni Design Technologies Advances 200G-Class Co-Packaged Optics IP Portfolio for Next-Generation AI Infrastructure
- Faraday Broadens IP Offerings on UMC’s 14nm Process for Edge AI and Consumer Markets
Latest News
- Hardware Root of Trust Essential for AI Chip Integrity
- AI Compute Demand Drives 44% YoY Growth for Top 10 Global Fabless IC Firms in 2025
- IBM Announces Strategic Collaboration with Arm to Shape the Future of Enterprise Computing
- Rambus Unveils HBM4E Controller: 16 GT/s, 2,048-Bit Interface, Enabling C-HBM4E
- AimFuture, a Leader in Home Appliance NPUs, to Integrate Mesacure Company’s AI Algorithms