2 ms·
Did you modify the lambda.cpp repo to move frmo 4-bit to 8-bit quantization? Or did you write something custom?
by dangoodmanUT 4y ago
Did you modify the lambda.cpp repo to move frmo 4-bit to 8-bit quantization? Or did you write something custom?