ggerganov/whisper.cpp

Pull Requests Commits

ci : use local ggml

ggerganov committed 1 year ago

Verified 511579cc

ggml : tmp workaround for whisper.cpp (skip) (#2565)

ggerganov committed 1 year ago

Verified bb12cd9b

update : readme

ggerganov committed 1 year ago

Verified f02b40bc

scripts : fix sync path

ggerganov committed 1 year ago

Verified 83ac2842

whisper.swiftui : switch Mac dest to Mac (Designed for iPad) (#2562)

jhen0409 committed 1 year ago

Verified c4e95fb7

cmake : fix ppc64 check (#0)

ggerganov committed 1 year ago

e23721f3

whisper : include ggml-cpu.h (#0)

ggerganov committed 1 year ago

c0a9f8ef

ggerganov committed 1 year ago

6477b84e

talk-llama : sync llama.cpp

ggerganov committed 1 year ago

24d70677

whisper : fix build (#0)

ggerganov committed 1 year ago

5089ab2d

ggerganov committed 1 year ago

bdbb9068

sycl : Fixes to broken builds and test-backend-ops (llama/10257)

Alberto Cabrera Pérez committed 1 year ago

fa2ebd33

vulkan: Optimize contiguous copies (llama/10254)

jeffbolznv committed 1 year ago

21b01a21

vulkan: Throttle the number of shader compiles during the build step. (llama/10222)

jeffbolznv committed 1 year ago

b54ce5ed

metal : more precise Q*K in FA vec kernel (llama/10247)

ggerganov committed 1 year ago

26a31b78

vulkan: Fix newly added tests for permuted mul_mat and 1D im2col (llama/10226)

jeffbolznv committed 1 year ago

14d13c5f

metal : reorder write loop in mul mat kernel + style (llama/10231)

ggerganov committed 1 year ago

5e110c2e

metal : fix build and some more comments (llama/10229)

ggerganov committed 1 year ago

4a9926d5

metal : fix F32 accumulation in FA vec kernel (llama/10232)

ggerganov committed 1 year ago

ae3c5642

metal : hide debug messages from normal log

ggerganov committed 1 year ago

e287a3b6

ggml: fix zero division in ‘dne’ calculation in CUDA COUNT_EQUAL operator when ‘ne’ is small (#10213)

SongXiaoXi committed 1 year ago

b8902436

ggml : optimize llamafile cpu matrix multiplication for ppc64le (llama/10156)

amritahs-ibm committed 1 year ago

b7b38f7d

metal : opt-in compile flag for BF16 (llama/10218)

ggerganov committed 1 year ago

9f67aab2

metal : improve clarity (minor) (llama/10171)

ggerganov committed 1 year ago

8f0f785d

metal : optimize FA kernels (llama/10171)

ggerganov committed 1 year ago

d0b83357

ggml : add ggml-cpu.h to the public headers (llama/10204)

slaren committed 1 year ago

1550be79

fix q4_0_8_8 format for corrupted tokens issue (llama/10198)

snadampal committed 1 year ago

807f848c

Optimize RWKV6 Operator Naming and Implement Multi-core CPU/ SYCL Acceleration (llama/10133)

Zhiyuan Li committed 1 year ago

42398f13

metal : add BF16 support (llama/8439)

ggerganov committed 1 year ago

31c3482a

metal : fix from ptr buffer name (llama/10189)

slaren committed 1 year ago

50257af6

Older