Merge pull request #203 from janhq/update-submodule-2024-08-25-17-07 · janhq/cortex.llamacpp@9eed833

Commit

Merge pull request #203 from janhq/update-submodule-2024-08-25-17-07

Update llama.cpp submodule to latest release b3620

jan-service-account authored Aug 25, 2024

2 parents 2e934ae + 4c5dc54 commit 9eed833

+10 −5		convert_hf_to_gguf.py
+1 −0		convert_lora_to_gguf.py
+1 −1		examples/quantize/quantize.cpp
+2 −1		ggml/include/ggml.h
+12 −5		ggml/src/ggml-cuda/fattn-common.cuh
+43 −9		ggml/src/ggml-cuda/fattn-tile-f16.cu
+40 −7		ggml/src/ggml-cuda/fattn-tile-f32.cu
+58 −13		ggml/src/ggml-cuda/fattn-vec-f16.cuh
+57 −11		ggml/src/ggml-cuda/fattn-vec-f32.cuh
+58 −5		ggml/src/ggml-cuda/fattn-wmma-f16.cuh
+2 −2		ggml/src/ggml-cuda/fattn.cu
+9 −0		ggml/src/ggml-metal.m
+23 −10		ggml/src/ggml.c
+2 −7		src/llama.cpp
+13 −9		tests/test-backend-ops.cpp
+1 −1		tests/test-lora-conversion-inference.sh