Download llama_cpp_python-0.3.10+cpuavx2-cp310-cp310-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/689e07b79aa36bc4d3ec6f9aacca5f2d032d110f/llama_cpp_python-0.3.10%2Bcpuavx2-cp310-cp310-linux_x86_64-basic.whl
- Command line
-
hf download hf://datasets/AIencoder/llama-cpp-wheels@689e07b79aa36bc4d3ec6f9aacca5f2d032d110f/llama_cpp_python-0.3.10+cpuavx2-cp310-cp310-linux_x86_64-basic.whl
-
curl -L -o llama_cpp_python-0.3.10+cpuavx2-cp310-cp310-linux_x86_64-basic.whl https://huggingface.co/datasets/AIencoder/llama-cpp-wheels/resolve/689e07b79aa36bc4d3ec6f9aacca5f2d032d110f/llama_cpp_python-0.3.10%2Bcpuavx2-cp310-cp310-linux_x86_64-basic.whl
4.29 MB
- Xet hash:
- fa748318c2305be27f473efd067e1aafbb4d1e488e07e41fafdc4a8af3447692
- Size of remote file:
- 4.29 MB
- SHA256:
- 7b916bc78e7c89bf1884de37a5b855e951d65d13c257cbe9c8dd78b229473203
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.