Releases: ggml-org/llama.cpp
Releases · ggml-org/llama.cpp
b4821
main: allow preloading conversation with -p and add -st / --single-tu…
b4820
`server`: fix deadly typo in response_format.json_schema.schema handl…
b4819
HIP: implement FlashAttention via rocWMMA for CDNA and RDNA3+ (#12032) Adds GGML_HIP_ROCWMMA_FATTN and rocwmma header check Adds rocWMMA support to fattn-wmma-f16 --- Signed-off-by: Carl Klemm <[email protected]> Co-authored-by: Johannes Gäßler <[email protected]> Co-authored-by: Ben Jackson <[email protected]>
b4818
sync : ggml ggml-ci
b4806
tts: add speaker file support (#12048) * tts: add speaker file support Signed-off-by: dm4 <[email protected]> * tts: handle outetts-0.3 * tts : add new line in error message --------- Signed-off-by: dm4 <[email protected]> Co-authored-by: Georgi Gerganov <[email protected]>
b4805
test-backend-ops : add option -p to filter by op params (#12155)
b4804
ggml : fix kleidiai build (#12159) The libggml API has changed, but this has not been updated.
b4803
Adding UTF-8 support to llama.cpp (#12111) For emojis, non-alpha characters, etc. Signed-off-by: Eric Curtin <[email protected]>
b4801
SYCL: Move CPY kernels to a separate file and add few missing kernels…
b4800
ggml-backend : keep paths in native string type when possible (#12144)