11 lines
229 B
Plaintext
11 lines
229 B
Plaintext
|
# inference (GGUF, grammar-constrained, recommended for the 4GB laptop)
|
||
|
|
llama-cpp-python>=0.2.79
|
||
|
|
|
||
|
|
# inference via the merged fp16 model (optional)
|
||
|
|
transformers>=4.44
|
||
|
|
accelerate>=0.30
|
||
|
|
torch>=2.1
|
||
|
|
|
||
|
|
# utilities
|
||
|
|
huggingface_hub>=0.23
|