top of page
Floating-Point Adders
"A bunch of tensors getting multiplied and added": this is what 70% of AI workloads boil down to. Nothing fancy. My passion for...
Mohamed Abdelgawad
Feb 11 min read
The Road to MX: The Evolution of AI Data Formats (INT8, Bfloat, FP8, MSFP, MX)
If there is one factor that has driven the most significant AI efficiency gains in the past decade, it is the evolution of data formats....
Mohamed Abdelgawad
Jan 117 min read
SRAM Alone: The Bandwidth King That Falls Short on Capacity for LLMs
When it comes to running large language models like LLaMA-3-70B, one fact stands out: while on-chip SRAM excels at providing high...
Mohamed Abdelgawad
Dec 20, 20244 min read
bottom of page