feat: Update llama.cpp · abetlen/llama-cpp-python@1615eb9

Commit

feat: Update llama.cpp

Browse files

abetlen committed Jun 10, 2024

1 parent 83d6b26 commit 1615eb9

+5 −0		.github/PULL_REQUEST_TEMPLATE/pull_request_template.md
+2 −4		.github/workflows/server.yml
+14 −0		CONTRIBUTING.md
+0 −29		README.md
+14 −12		convert-hf-to-gguf.py
+0 −19		examples/alpaca.sh
+0 −15		examples/gpt4all.sh
+51 −7		examples/imatrix/imatrix.cpp
+0 −18		examples/llama2-13b.sh
+0 −18		examples/llama2.sh
+1 −1		examples/server/public/index-new.html
+16 −14		examples/server/server.cpp
+3 −3		flake.lock
+57 −31		ggml-cuda.cu
+17 −4		ggml-cuda/common.cuh
+10 −10		ggml-cuda/fattn-common.cuh
+1 −1		ggml-cuda/fattn-tile-f16.cu
+1 −1		ggml-cuda/fattn-vec-f16.cuh
+3 −3		ggml-cuda/fattn-wmma-f16.cuh
+95 −0		ggml-cuda/mma.cuh
+2 −1		ggml-cuda/mmq.cu
+548 −139		ggml-cuda/mmq.cuh
+77 −10		ggml-cuda/quantize.cu
+16 −1		ggml-cuda/quantize.cuh
+7 −4		ggml-sycl.cpp