whisper.cpp

Running

App Files Files Community

whisper.cpp / ggml /src

6.6 MB

100 contributors

History: 553 commits

cmdr2

Support pure float16 add/sub/mul/div operations in the CUDA (and CPU) backend (ggml/1121)

2b94a24 about 1 year ago

ggml-amx
ggml : adapt AMX to tensor->grad removal (llama/0) over 1 year ago
ggml-blas
ggml : add support for dynamic loading of backends (llama/10469) over 1 year ago
ggml-cann
llama : add Qwen2VL support + multimodal RoPE (llama/10361) about 1 year ago
ggml-cpu
Support pure float16 add/sub/mul/div operations in the CUDA (and CPU) backend (ggml/1121) about 1 year ago
ggml-cuda
Support pure float16 add/sub/mul/div operations in the CUDA (and CPU) backend (ggml/1121) about 1 year ago
ggml-hip
CUDA: app option to compile without FlashAttention (llama/12025) about 1 year ago
ggml-kompute
llama : add Qwen2VL support + multimodal RoPE (llama/10361) about 1 year ago
ggml-metal
metal : copy kernels for quant to F32/F16 conversions (llama/12017) about 1 year ago
ggml-musa
CUDA: app option to compile without FlashAttention (llama/12025) about 1 year ago
ggml-opencl
opencl: fix for small models (llama/11950) about 1 year ago
ggml-rpc
rpc: fix known RCE in rpc-server (ggml/1103) about 1 year ago
ggml-sycl
Optimize mul_mat for Q4_0 on Intel GPU (llama/12035) about 1 year ago
ggml-vulkan
vulkan: implement several ops relevant for ggml_opt (llama/11769) about 1 year ago
CMakeLists.txt

11.8 kB

`ci`: use sccache on windows instead of ccache (llama/11545) about 1 year ago
ggml-alloc.c

38.1 kB

vulkan: use smaller combined allocations to avoid fragmentation (llama/11551) about 1 year ago
ggml-backend-impl.h

12 kB

rpc : early register backend devices (llama/11262) about 1 year ago
ggml-backend-reg.cpp

17.2 kB

ggml : allow loading backend with env variable (ggml/1059) about 1 year ago
ggml-backend.cpp

77.5 kB

ggml-backend : only offload from host buffers (fix) (llama/11124) about 1 year ago
ggml-common.h

133 kB

CUDA: use arch list for compatibility check (llama/11775) about 1 year ago
ggml-impl.h

18.4 kB

MUSA: support ARM64 and enable dp4a .etc (llama/11843) about 1 year ago
ggml-opt.cpp

31.7 kB

ggml-opt: fix data corruption (ggml/1022) over 1 year ago
ggml-quants.c

214 kB

ggml : refactor online repacking (llama/10446) over 1 year ago
ggml-quants.h

8.34 kB

ggml : build backends as libraries (llama/10256) over 1 year ago
ggml-threading.cpp

250 Bytes

ggml : build backends as libraries (llama/10256) over 1 year ago
ggml-threading.h

198 Bytes

remove CMAKE_WINDOWS_EXPORT_ALL_SYMBOLS (llama/10797) about 1 year ago
ggml.c

209 kB

ggml-cpu: Support s390x SIMD Instruction Set (llama/12019) about 1 year ago
gguf.cpp

45 kB

cmake : add sanitizer flags for llama.cpp (llama/11279) about 1 year ago