Merge pull request #141 from janhq/update-submodule-2024-07-12-17-05 · janhq/cortex.llamacpp@861a591

Commit

Merge pull request #141 from janhq/update-submodule-2024-07-12-17-05

Update llama.cpp submodule to latest release b3384

jan-service-account authored Jul 12, 2024

2 parents fff9b0c + 0f76113 commit 861a591

+1 −1		.devops/tools.sh
+7 −1		Makefile
+1 −2		convert_hf_to_gguf.py
+1 −1		examples/eval-callback/eval-callback.cpp
+7 −7		examples/gguf-hash/gguf-hash.cpp
+7 −2		examples/main/main.cpp
+1 −1		examples/quantize-stats/quantize-stats.cpp
+32 −6		examples/server/server.cpp
+20 −2		examples/server/utils.hpp
+25 −25		ggml/include/ggml.h
+36 −32		ggml/src/ggml-aarch64.c
+7 −7		ggml/src/ggml-aarch64.h
+1 −1		ggml/src/ggml-backend.c
+9 −4		ggml/src/ggml-blas.cpp
+1 −1		ggml/src/ggml-cuda/common.cuh
+47 −47		ggml/src/ggml-quants.c
+19 −19		ggml/src/ggml-quants.h
+13 −36		ggml/src/ggml-sycl.cpp
+55 −56		ggml/src/ggml.c
+1 −8		src/llama.cpp
+2 −2		tests/test-double-float.cpp
+1 −1		tests/test-quantize-fns.cpp
+1 −1		tests/test-quantize-perf.cpp