ggerganov/whisper.cpp

Pull Requests Commits

benches : update

ggerganov committed 67 days ago

Verified 4bbce1e5

ggerganov committed 67 days ago

f5b477ab

ggml : bump version to 0.9.8 (ggml/1442)

ggerganov committed 67 days ago

b2be1620

ggml : restore ggml_type_sizef() to aboid major version bump (ggml/1441)

ggerganov committed 67 days ago

945d3151

fix: VAD time mapping timestamp drift caused by overlap samples (#3711)

lohopupa committed 68 days ago

Verified dc961166

go : handle EOF correctly in model download (#3671)

Lumberj3ck committed 69 days ago

Verified 79218f51

py : replace deprecated openvino-dev with openvino>=2023.3.0 (#3678)

Aiudadadadf committed 69 days ago

Verified 975b9798

examples : Allow max_len to be used for any output format (#3679)

gaelj committed 69 days ago

Verified 21665eab

server: return proper HTTP status codes for error responses (#3707)

dearlordylord committed 69 days ago

Verified 136dc2eb

ggml : try fix arm build (#0)

ggerganov committed 69 days ago

27fa2077

talk-llama : sync llama.cpp

ggerganov committed 69 days ago

2bc630f1

ggerganov committed 69 days ago

ab1252c1

ggml : extend im2col f16 (ggml/1434)

David366AI committed 69 days ago

d4bc3121

common : add nvfp4 (ggml/0)

ggerganov committed 69 days ago

81ea9587

CUDA: limit number of FA stream-k CUDA blocks (llama/20586)

JohannesGaessler committed 69 days ago

d7926e62

ggml: avoid creating CUDA context during device init (llama/20595)

ServeurpersoCom committed 69 days ago

2fb6aea8

ggml/hip: fix APU compatibility - soft error handling for hipMemAdviseSetCoarseGrain (llama/20536)

moonshadow-25 committed 69 days ago

b327a321

ggml : guard against sumq2 being 0 in IQ4_NL (llama/20460)

bartowski1182 committed 69 days ago

67702398

cuda : add RDNA4-specific MMVQ parameter table for bs=1 decode (llama/19478)

JoursBleu committed 69 days ago

55c66106

vulkan: use graphics queue on AMD (llama/20551)

0cc4m committed 69 days ago

cd02195b

metal : add FA specialization for HSK = 320, HSV = 256 (llama/20549)

ggerganov committed 69 days ago

b3120184

hexagon: Q4_0 and MXFP4 repack fixes (llama/20527)

max-krasnyansky committed 69 days ago

55f8cfda

add op gated_delta_net (llama/20455)

arthw committed 69 days ago

c5f9a49b

ggml : add native AVX512-FP16 support for F16 operations (llama/20529)

angt committed 69 days ago

93d09fdb

Use fp32 in cuBLAS V100 to avoid overflows, env variables to override cuBLAS compute type (llama/19959)

wallentri88 committed 69 days ago

8ad5cb1e

ggml : add OpenVINO backend (llama/15307)

wine99 committed 69 days ago

96b163e8

Fix data race in CUDA's "cpy" kernel (influences GGML's DUP, CONT operations). (llama/20507)

Exile333 committed 69 days ago

46aad766

opencl: fix l2_norm (llama/20480)

lhez committed 69 days ago

a31600d8

graph : remove redundant GDN state transposes (llama/20443)

ggerganov committed 69 days ago

c7abcd57

ggml-cpu: add RVV vec dot kernels for quantization types (llama/18859)

rehan-10xengineer committed 69 days ago

5905e870

Older