Quantize with newest llama.cpp

c4ai-command-r-v01-imat-Q3_K_S.gguf

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c3d5e660fd12fca136996072d3d2234bfa1315686d587ef9142b4bdc75ca9c45
-size 15862110016
+oid sha256:1ed6e4ae399bc44f346f667748e403f75fd5c521fdab90c5264bc7fdcbc3fcd4
+size 15862110304