]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: d5fe4e8)
ggml: move fp16/bf16 conversion optimizations to CPU backend + export conversion...
authorSXX <redacted>
Sat, 26 Apr 2025 14:05:31 +0000 (22:05 +0800)
committerGitHub <redacted>
Sat, 26 Apr 2025 14:05:31 +0000 (16:05 +0200)
commit77d5e9a76a7b4a8a7c5bf9cf6ebef91860123cba
tree3fe29c6375fd35f2a9c1333224d4beda22b86f8ftree
parentd5fe4e81bd447124836ecfb47d794f8768665b9fcommit | diff
ggml: move fp16/bf16 conversion optimizations to CPU backend + export conversion APIs (#13107)

* ggml: dynamic x86_64 feature detection for FP32 <-> FP16/BF16 conversion

* move fp converter to ggml-cpu

* Switch ggml_compute_forward_get_rows_f16/bf16 to new ggml_cpu_fp16/bf16_to_fp32
ggml/include/ggml-cpu.h diff | blob | history
ggml/src/ggml-cpu/ggml-cpu.c diff | blob | history
ggml/src/ggml-cpu/ops.cpp diff | blob | history
ggml/src/ggml.c diff | blob | history
Packaging of ggml-org/llama.cpp