llama.cpp updated · mgonzs13/llama_ros@27d01d2 · GitHub

Commit

llama.cpp updated

Browse files

Loading branch information

mgonzs13 committed Feb 27, 2024

1 parent 0353b70 commit 27d01d2

llama_ros/llama_cpp

Submodule llama_cpp updated 21 files

+17 −61		.github/workflows/server.yml
+7 −2		Makefile
+1 −0		README.md
+9 −0		common/common.cpp
+1 −0		common/common.h
+2 −2		examples/passkey/passkey.cpp
+7 −3		examples/quantize/quantize.cpp
+4 −0		examples/server/server.cpp
+502 −54		ggml-cuda.cu
+58 −8		ggml-metal.m
+708 −3		ggml-metal.metal
+1,012 −30		ggml-quants.c
+27 −0		ggml-quants.h
+165 −81		ggml-sycl.cpp
+61 −0		ggml.c
+4 −0		ggml.h
+136 −56		llama.cpp
+5 −1		llama.h
+2 −2		tests/test-backend-ops.cpp
+3 −1		tests/test-quantize-fns.cpp
+2 −1		unicode.h

0 comments on commit `27d01d2`

Please sign in to comment.