ggml-org/llama.cpp

Pull Requests Commits

vulkan: support set_rows with i32 index type (#16162)

jeffbolznv committed 2 hours ago

Verified d8abd2d5

indent++ [no-ci]

CISC committed 2 hours ago

Verified bd47cba7

deduplicate cuda/sycl and test-fix [no-ci]

CISC committed 2 hours ago

Verified aa76826c

forgotten name change

CISC committed 6 hours ago

Verified 98991820

Apply suggestions from code review

CISC committed 9 hours ago

Verified 480b404d

test quantized path

CISC committed 10 hours ago

Verified 7657ec3d

CISC committed 18 hours ago

Verified 9a7bff30

implement set_rows with i32 index

CISC committed 19 hours ago

Verified a7403b13

ci : add label for the RISC-V runner (#16150)

ggerganov committed 1 day ago

Verified da30ab5f

ci : migrate ggml ci to self-hosted runners (#16116)

ggerganov committed 1 day ago

Verified 28baac9c

vulkan: optimize UMA buffer operations and fix driver hangs (#16059)

giuseppe committed 1 day ago

Verified 1eeb523c

vulkan: fix validation error about VK_PIPELINE_CREATE_CAPTURE_STATISTICS_BIT_KHR (#16086)

jeffbolznv committed 1 day ago

Verified 5bb4a3ed

ggerganov committed 2 days ago

7f766929

ggml : introduce semantic versioning (ggml/1336)

danbev committed 2 days ago

405921dc

CUDA : conditionally add cuda architectures (ggml/1341)

gjasny committed 2 days ago

fa6383ca

vulkan: use vec dot for matrix matrix multiplications (#16056)

0cc4m committed 2 days ago

Verified 803dac2e

server: fix SSE and OpenAI compatibility for error messages when streaming (#16109)

BenjaminBruenau committed 2 days ago

Verified 459c0c2c

llama-bench: add --devices and --list-devices support (#16039)

ssweens committed 2 days ago

Verified be79d9fd

chat: Fix streaming parser for granite models (#15682)

shun095 committed 3 days ago

Verified f432d8d8

feat: Improve mobile UI for Settings Dialog (#16084)

allozaur committed 3 days ago

Verified 4067f07f

chat : fix build on arm64 (#16101)

ngxson committed 3 days ago

Verified 4b8560ab

ggml : refactor forward_dup for cpu backend (#16062)

ngxson committed 3 days ago

Verified 0dd58b68

ggml-amx : fix ggml_amx_init() on generic Linux (#16049)

angt committed 3 days ago

Verified 69ffd891

cmake : fix static linking for OpenMP on Unix-like systems (#16031)

angt committed 3 days ago

Verified 246c0d9c

opencl: optimize mxfp4 kernels (#16037)

shawngu-quic committed 3 days ago

Verified 3edd87cd

rename optimize_graph to graph_optimize (#16082)

jeffbolznv committed 3 days ago

Verified c0b45097

CUDA: Optimize PAD_REFLECT_1D (#15957)

bugparty committed 3 days ago

Verified 38dbdf4c

CUDA: fix compilation on CC 6.0 (#16091)

JohannesGaessler committed 3 days ago

Verified 368560a1

Add resumable downloads for llama-server model loading (#15963)

ericcurtin committed 4 days ago

Verified 4ca088b0

metal : use function constants for mul_mv_ext kernels (#16074)

ggerganov committed 4 days ago

Verified 703f9e32

Older