Vector Post-Training Quantization (VPTQ) is a novel Post-Training Quantization method that leverages Vector Quantization to high accuracy on LLMs at an extremely low bit-width (<2-bit). VPTQ can ...
We want all inserts\extracts allocas to be 32bit for bools going into matrices like we did for vectors. We should use ConvertTypeForMem when we get a bool element type and then after that the storage ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback