0ceeb11721
* Merging mainline - WIP * Merging mainline - WIP AVX2 and CUDA appear to work. CUDA performance seems slightly (~1-2%) lower as it is so often the case with llama.cpp/ggml after some "improvements" have been made. * Merging mainline - fix Metal * Remove check --------- Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
6 lines
99 B
Plaintext
6 lines
99 B
Plaintext
numpy~=1.26.4
|
|
sentencepiece~=0.2.0
|
|
transformers>=4.40.1,<5.0.0
|
|
gguf>=0.1.0
|
|
protobuf>=4.21.0,<5.0.0
|