Download llama_cpp_python-0.3.10+cpuavx-cp313-cp313-linux_x86_64-basic.whl from AIencoder/llama-cpp-wheels: direct link, hf CLI and curl.
- Browser
- Download file 4.29 MB
-
https://huggingface.co/datasets/AIencoder/llama-cpp-wheels/resolve/fdeda035d3fd012c53b16da27dfdfd751d073083/llama_cpp_python-0.3.10%2Bcpuavx-cp313-cp313-linux_x86_64-basic.whl
- Command line
-
hf download hf://datasets/AIencoder/llama-cpp-wheels@fdeda035d3fd012c53b16da27dfdfd751d073083/llama_cpp_python-0.3.10+cpuavx-cp313-cp313-linux_x86_64-basic.whl
-
curl -L -o llama_cpp_python-0.3.10+cpuavx-cp313-cp313-linux_x86_64-basic.whl https://huggingface.co/datasets/AIencoder/llama-cpp-wheels/resolve/fdeda035d3fd012c53b16da27dfdfd751d073083/llama_cpp_python-0.3.10%2Bcpuavx-cp313-cp313-linux_x86_64-basic.whl
4.29 MB
- Xet hash:
- afa556edffb617e04fb6c38c7a1d35dd4b6bebaa9b364ee592116f2966700466
- Size of remote file:
- 4.29 MB
- SHA256:
- 6c6fafbfacb4d8528cabbc32807d75ae50b0ea6906ffcbe52d90e7a9e28769d8
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.