Merge pull request #275 from janhq/update-submodule-2024-10-30-17-07 · janhq/cortex.llamacpp@1191e7d

Commit

Merge pull request #275 from janhq/update-submodule-2024-10-30-17-07

Update llama.cpp submodule to latest release b3995

jan-service-account authored Oct 30, 2024

2 parents e3cfa42 + 2b59d01 commit 1191e7d

+3 −3		convert_lora_to_gguf.py
+22 −110		examples/llama-bench/llama-bench.cpp
+9 −2		examples/main/README.md
+12 −7		ggml/include/ggml-backend.h
+1 −1		ggml/include/ggml-cuda.h
+4 −0		ggml/include/ggml-kompute.h
+268 −0		ggml/src/ggml-aarch64.c
+8 −25		ggml/src/ggml-amx.cpp
+4 −15		ggml/src/ggml-backend-impl.h
+127 −115		ggml/src/ggml-backend.cpp
+5 −15		ggml/src/ggml-blas.cpp
+6 −44		ggml/src/ggml-cann.cpp
+63 −77		ggml/src/ggml-cuda.cu
+196 −72		ggml/src/ggml-kompute.cpp
+25 −19		ggml/src/ggml-metal.m
+2 −18		ggml/src/ggml-rpc.cpp
+14 −40		ggml/src/ggml-sycl.cpp
+8 −18		ggml/src/ggml-vulkan.cpp
+57 −14		ggml/src/ggml.c
+2 −5		include/llama.h
+1 −1		scripts/compare-llama-bench.py
+1,617 −1,505		src/llama.cpp