llama.cpp
Adding IQ2_S and IQ2_M to complete coverage of the 2-3 bit quantization range
#5721
Merged
Merged
Adding IQ2_S and IQ2_M to complete coverage of the 2-3 bit quantization range #5721
ikawrakow
merged
2 commits into
master
from
ik/iq2_s_new2
Adding IQ2_S and IQ2_M as a single cumulative commit
88a18f9f
ggerganov
approved these changes
on 2024-02-26
Update examples/quantize/quantize.cpp
ec0abd2d
Reviewers
Assignees
No one assigned
Labels
Review Complexity : High
Tensor Encoding Scheme
Milestone
No milestone
Login to write a write a comment.
Login via GitHub