Concept

TaSQ

A KV-cache quantization method for standard LLMs that can be stacked with LoopVQ to achieve quantization at lower bit rates.

Timeline (last 90 days)

Nov 4

When combined with TaSQ, LoopVQ retains most of BF16 precision even at an extremely low bit rate of 0.25 bits per element.

Unverified50%

When combined with TaSQ, LoopVQ retains most of BF16 precision even at an extremely low bit rate of 0.25 bits per element.