build: update llama.cpp · Adriankhl/godot-llm@ad66a77

Commit

build: update llama.cpp

Browse files

Adriankhl committed May 28, 2024

1 parent 9aba11e commit ad66a77

llama.cpp

Submodule llama.cpp updated 27 files

+1 −1		.github/ISSUE_TEMPLATE/05-enhancement.yml
+1 −1		.github/ISSUE_TEMPLATE/06-question.yml
+28 −0		.github/ISSUE_TEMPLATE/07-refactor.yml
+4 −0		CMakeLists.txt
+6 −2		CMakePresets.json
+3 −0		Makefile
+2 −1		README.md
+92 −2		convert-hf-to-gguf.py
+1 −1		examples/llava/clip.h
+26 −15		examples/server/public/index.html
+24 −9		ggml-cuda.cu
+0 −5		ggml-cuda/common.cuh
+87 −6		ggml-cuda/concat.cu
+61 −10		ggml-metal.m
+63 −17		ggml-metal.metal
+75 −58		ggml-rpc.cpp
+337 −159		ggml-sycl.cpp
+40 −23		ggml.c
+3 −2		ggml.h
+63 −11		gguf-py/gguf/constants.py
+21 −0		gguf-py/gguf/gguf_writer.py
+28 −1		gguf-py/gguf/tensor_mapping.py
+501 −181		llama.cpp
+5 −3		llama.h
+17 −11		tests/test-backend-ops.cpp
+2 −0		tests/test-tokenizer-0.sh
+13 −7		tests/test-tokenizer-random.py

Please sign in to comment.