sync : ggml#3108

Merged

ggerganov merged 16 commits intomasterfrom

sync-ggml-25-05-01

May 1, 2025

Commits on May 1, 2025

ggml : fix ggml_gallocr_ptr type (ggml/1205)

slaren
authored and
ggerganov
committed
rpc : do not wait for response when sending RPC_CMD_SET_TENSOR (llama/12943)

rgerganov
authored and
ggerganov
committed
change the reorder tensor from init to execute OP (llama/13003)

NeoZhangJianyu
authored and
ggerganov
committed
ggml: move fp16/bf16 conversion optimizations to CPU backend + export conversion APIs (llama/13107)

SongXiaoXi
authored and
ggerganov
committed
musa: fix build warning (llama/13129)

yeahdongcn
authored and
ggerganov
committed
CUDA: fix q_nope_absorbed prec for DS 2 Lite f16 (llama/13137)

JohannesGaessler
authored and
ggerganov
committed
musa: fix typo in cc control (llama/13144)

yeahdongcn
authored and
ggerganov
committed
SYCL: Add all missing unary kernels (llama/13074)

qnixsynapse
authored and
ggerganov
committed
fix(rpc): Improve input validation and error handling (llama/13069)

thevilledev
authored and
ggerganov
committed
CUDA: fix non-cont. inputs for batched mat mul (llama/13155)

JohannesGaessler
authored and
ggerganov
committed
feat(ggml-cpu): enable z17 compile (llama/13182)

taronaeo
authored and
ggerganov
committed
ggml : fix ppc64le build (llama/13176)

shalinib-ibm
authored and
ggerganov
committed
vulkan: use uint array index to avoid glslang bug (llama/13193)

jeffbolznv
authored and
ggerganov
committed
CUDA: batched+noncont MMQ, refactor bs>1 MoE code (llama/13199)

JohannesGaessler
authored and
ggerganov
committed
sync : ggml
ggerganov
committed
talk-llama : sync llama.cpp
ggerganov
committed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

sync : ggml#3108

sync : ggml#3108
ggerganov merged 16 commits intomasterfrom
sync-ggml-25-05-01

Commits on May 1, 2025

ggml : fix ggml_gallocr_ptr type (ggml/1205)

rpc : do not wait for response when sending RPC_CMD_SET_TENSOR (llama/12943)

change the reorder tensor from init to execute OP (llama/13003)

ggml: move fp16/bf16 conversion optimizations to CPU backend + export conversion APIs (llama/13107)

musa: fix build warning (llama/13129)

CUDA: fix q_nope_absorbed prec for DS 2 Lite f16 (llama/13137)

musa: fix typo in cc control (llama/13144)

SYCL: Add all missing unary kernels (llama/13074)

fix(rpc): Improve input validation and error handling (llama/13069)

CUDA: fix non-cont. inputs for batched mat mul (llama/13155)

feat(ggml-cpu): enable z17 compile (llama/13182)

ggml : fix ppc64le build (llama/13176)

vulkan: use uint array index to avoid glslang bug (llama/13193)

CUDA: batched+noncont MMQ, refactor bs>1 MoE code (llama/13199)

sync : ggml

talk-llama : sync llama.cpp

sync : ggml#3108ggerganov merged 16 commits intomasterggml-org/whisper.cpp:masterfrom sync-ggml-25-05-01ggml-org/whisper.cpp:sync-ggml-25-05-01Copy head branch name to clipboard

Commits

Commits on May 1, 2025

sync : ggml#3108
ggerganov merged 16 commits intomasterfrom
sync-ggml-25-05-01