public_git/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/whisper.cpp.git synced 2026-10-10 08:15:40 +02:00
Code Issues Packages Projects Releases Wiki Activity
Files
fadb3233b6dcf86f6b5e998dbabb84e43d78e0d7
whisper.cpp/ggml/src
T
History
Jeff Bolz fadb3233b6 vulkan: optimize flash attention split_k_reduce (llama/14554) 
* vulkan: allow FA split_k with smaller KV values

* vulkan: spread split_k_reduce work across more threads

k_num can get rather large. Use the whole workgroup to reduce the M/L values.

Launch a thread for each element in the HSV dimension of the output. Helps a
lot for large HSV (like deepseek).
2025-07-12 19:23:56 +03:00
..
ggml-amx
…
ggml-blas
cmake : Fix broken CMake error messages (ggml/1252)
2025-06-01 15:14:44 +03:00
ggml-cann
CANN: Replace aclrtMemsetSync with aclnnInplaceZero operator (llama/14002)
2025-07-12 19:23:56 +03:00
ggml-cpu
ggml : implement GEGLU_ERF and GEGLU_QUICK ops (llama/14445)
2025-07-12 19:23:56 +03:00
ggml-cuda
cuda : fix rope with partial rotation and non-cont src (llama/14580)
2025-07-12 19:23:56 +03:00
ggml-hip
HIP: disable rocwmma on gfx12 by default until rocm 7.0 (llama/14202)
2025-06-18 12:40:34 +03:00
ggml-kompute
…
ggml-metal
metal : disable fast math in all quantize kernels (llama/14528)
2025-07-12 19:23:56 +03:00
ggml-musa
musa: enable fp16 mma (all) and cublas on qy2 (llama/13842)
2025-07-01 17:54:53 +03:00
ggml-opencl
opencl: add GELU_ERF (llama/14476)
2025-07-12 19:23:56 +03:00
ggml-rpc
rpc : nicer error messages for RPC server crash (llama/14076)
2025-06-18 12:40:34 +03:00
ggml-sycl
cuda : fix rope with partial rotation and non-cont src (llama/14580)
2025-07-12 19:23:56 +03:00
ggml-vulkan
vulkan: optimize flash attention split_k_reduce (llama/14554)
2025-07-12 19:23:56 +03:00
CMakeLists.txt
Implement GGML_CPU_ALL_VARIANTS for PowerPC (llama/14286)
2025-06-21 07:34:17 +03:00
ggml-alloc.c
ggml: Don't assert fail when tensor data changes (llama/13222)
2025-05-07 15:39:32 +03:00
ggml-backend-impl.h
…
ggml-backend-reg.cpp
build : suppress gcc15 compile warnings (llama/14261)
2025-06-21 07:34:17 +03:00
ggml-backend.cpp
vulkan: Add fusion support for RMS_NORM+MUL (llama/14366)
2025-07-01 17:54:53 +03:00
ggml-common.h
ggml-cpu : split arch-specific implementations (llama/13892)
2025-06-10 12:40:33 +03:00
ggml-impl.h
vulkan: Add fusion support for RMS_NORM+MUL (llama/14366)
2025-07-01 17:54:53 +03:00
ggml-opt.cpp
mnist: fix segmentation fault (ggml/1227)
2025-05-19 14:58:39 +03:00
ggml-quants.c
ggml-quants : rename best_mad to best_error (ggml/1283)
2025-07-01 17:54:53 +03:00
ggml-quants.h
…
ggml-threading.cpp
…
ggml-threading.h
…
ggml.c
ggml : implement GEGLU_ERF and GEGLU_QUICK ops (llama/14445)
2025-07-12 19:23:56 +03:00
ggml.cpp
ggml : Print backtrace on uncaught C++ exceptions (ggml/1232)
2025-05-29 09:56:26 +03:00
gguf.cpp
ggml : do not output unprintable characters on GGUF load failure (llama/14381)
2025-07-01 17:54:53 +03:00