ggerganov/whisper.cpp

Pull Requests Commits

ci : add ignore for bindings/{ruby, go} in build.yml [no ci] (#3837)

danbev committed 1 hour ago

Verified c932729a

ci : fix include paths for bindings-go job [no ci] (#3835)

danbev committed 4 hours ago

Verified e47a3eeb

ci : add on push/pull_request paths ruby job (#3833)

danbev committed 4 hours ago

Verified f41562bd

ci : renable arm64 docker builds (#3832)

danbev committed 7 hours ago

Verified 9186e245

ci : set GGML_NATIVE=OFF for bindings-java (#3830)

danbev committed 12 hours ago

Verified f6e617ba

ci : only run docker jobs when pushed to master [no ci] (#3828)

danbev committed 1 day ago

Verified 6dcdd653

docs : add AGENTS.md and CONTRIBUTING.md [no ci] (#3826)

danbev committed 1 day ago

Verified ee540bf0

cli : merge tokens split across UTF-8 boundaries in JSON output (#3751)

texasich committed 2 days ago

Verified 27101c01

release : v1.8.5

ggerganov committed 3 days ago

Verified e0fd1f67

benches : update

ggerganov committed 3 days ago

Verified c245b3ec

ggerganov committed 3 days ago

Verified f14ae77f

ggml : bump version to 0.13.0 (ggml/1510)

ggerganov committed 3 days ago

Verified 1cf8e3a9

TP: fix ggml context size calculation (llama/22616)

JohannesGaessler committed 3 days ago

Verified bcff5151

ggml: `gguf_init_from_callback` and `gguf_init_from_buffer` (llama/22341)

giladgd committed 3 days ago

Verified 2979e5f9

readme : add AMD ROCm/HIP GPU build instructions (#3823)

Kaihui-AMD committed 3 days ago

Verified 44a50ca4

talk-llama : sync llama.cpp

ggerganov committed 3 days ago

865ec171

ggerganov committed 3 days ago

0a62a579

ggml : bump version to 0.12.1 (ggml/1508)

ggerganov committed 3 days ago

946d6813

ggml : Parallelize quant LUT init (llama/23595)

jeffbolznv committed 3 days ago

a369b394

TP: fix entirely zero-sized slices per device (llama/23525)

JohannesGaessler committed 3 days ago

3306af62

opencl: batch profiling to improve speed and prevent memory leaks (llama/23495)

shaofeiqi committed 3 days ago

1435988a

hexagon: apply repl optimization in flash attn softmax as #22993 (llama/23455)

njsyw1997 committed 3 days ago

b84d0348

ggml : Check the right iface method before using the fallback 2d get (llama/23514)

dskwe committed 3 days ago

511f8602

vulkan: fix windows find_package of SPIRV-Headers (llama/23215)

jeffbolznv committed 3 days ago

6b85d73b

opencl: generalize Adreno MoE kernels on M (llama/23449)

shawngu-quic committed 3 days ago

aefffa1f

SYCL: improve MoE prefill throughput (llama/23142)

sanmai committed 3 days ago

b0c9f900

sycl : Level Zero detection in ggml_sycl_init (llama/23097)

sanmai committed 3 days ago

21c65a78

SYCL : gated_delta_net K>1 (llama/23174)

karavayev committed 3 days ago

0416feec

SYCL: add BF16 to DMMV kernel path (~4x tg speedup on Intel Arc) (llama/21580)

PMZFX committed 3 days ago

6fb7f1af

ggml-zendnn : add Q8_0 quantization support (llama/23414)

z-sachin committed 3 days ago

2d629533

Older