factual

Laziness Fuses Kernels and Cuts Memory Access

Tinygrad uses laziness like Haskell: instead of computing A times B immediately and storing it to memory, it waits until the user realizes the tensor, allowing the plus C operation to be fused into one kernel, which eliminates the excessive loads and stores to memory that PyTorch incurs.

factualpending

Speaker

George Hotz

Evidence Quote

“if you wait until the user actually realizes that tensor, until the laziness actually resolves, you can fuse that plus C.”

Source

George Hotz #3— Lex Fridman Podcast
Created: 6/13/2026, 12:26:10 AM

Your notes

Loading notes...