Fetching metadata from the HF Docker repository...
release : v1.4.3
4d80650
unverified
-
.devops
docker : Add dockerfile for cublas (#1286)
-
.github
ci : fix variable names in GitHub actions config (#1440)
-
bindings
release : v1.4.3
-
cmake
cmake : update to 3.19 (#351)
-
coreml
whisper : Metal and ggml-alloc support (#1270)
-
examples
whisper : add support for large v3 (#1444)
-
extra
whisper : add support for large v3 (#1444)
-
models
whisper : add support for large v3 (#1444)
-
openvino
whisper : add OpenVINO support (#1037)
-
samples
Create README.md
-
tests
whisper : add support for large v3 (#1444)
-
804 Bytes
Initial release
-
696 Bytes
extra: Add benchmark script implemented in Python (#1298)
-
96 Bytes
cmake : add submodule whisper.spm
-
18.3 kB
release : v1.4.3
-
1.07 kB
license : update year (#739)
-
14.6 kB
whisper : add support for large v3 (#1444)
-
36 kB
release : v1.4.3
-
27.1 kB
sync : ggml (backend v2, k-quants, CUDA opts, Metal opts, etc.) (#1422)
-
3.47 kB
sync : ggml (backend v2, k-quants, CUDA opts, Metal opts, etc.) (#1422)
-
3.27 kB
sync : ggml (backend v2, k-quants, CUDA opts, Metal opts, etc.) (#1422)
-
36.3 kB
sync : ggml (backend v2, k-quants, CUDA opts, Metal opts, etc.) (#1422)
-
5.89 kB
sync : ggml (backend v2, k-quants, CUDA opts, Metal opts, etc.) (#1422)
-
312 kB
cuda : fix HIPBLAS build
-
1.92 kB
sync : ggml (backend v2, k-quants, CUDA opts, Metal opts, etc.) (#1422)
-
7.38 kB
sync : ggml (backend v2, k-quants, CUDA opts, Metal opts, etc.) (#1422)
-
3.88 kB
sync : ggml (backend v2, k-quants, CUDA opts, Metal opts, etc.) (#1422)
-
86.3 kB
metal : fix asserts for setThreadgroupMemoryLength (close #1435)
-
103 kB
sync : ggml (backend v2, k-quants, CUDA opts, Metal opts, etc.) (#1422)
-
71 kB
sync : ggml (backend v2, k-quants, CUDA opts, Metal opts, etc.) (#1422)
-
845 Bytes
ggml : sync latest ggml lib
-
291 kB
sync : ggml (backend v2, k-quants, CUDA opts, Metal opts, etc.) (#1422)
-
10.2 kB
sync : ggml (backend v2, k-quants, CUDA opts, Metal opts, etc.) (#1422)
-
658 kB
ggml : fix MIN / MAX macro re-definition
-
78.8 kB
sync : ggml (backend v2, k-quants, CUDA opts, Metal opts, etc.) (#1422)
-
209 kB
whisper : add support for large v3 (#1444)
-
28.5 kB
whisper : add support for large v3 (#1444)
