Tag: quantization
All the articles with the tag "quantization".
-
Trading MatMuls for SRAM Lookups: A 3-Bit Edge Architecture
By trading heavy FP16 MatMuls for SRAM lookups and 1-bit additions, our custom quantization pipeline squeezes state-of-the-art models down to approx. 3 bits per weight with minimal accuracy loss. Here is how bypassing Tensor Cores could reshape the design of future edge AI chips.