John Shaughnessy
John Shaughnessy
Thanks to Porfirio for providing the screenshots and documenting the issue he encountered!
I am running into new errors that I can't explain. I removed the `llamafile` repo entirely and recloned it. When I ran `make -j32`, I encountered: ``` 32 errors and...
Setting `export CUDA_VISIBLE_DEVICES=1` works for me. I'll just use one GPU until multiple device support is fixed in llama.cpp. Thank you! The other issues that I started encountering (in this...