Quantized GGUF available

#3
by MaziyarPanahi - opened

Hi,

Thanks for sharing your model. I quantized it to GGUF models for those with low resources: https://huggingface.co/MaziyarPanahi/luxia-21.4b-alignment-v1.0-GGUF

Thanks again

GGML_ASSERT: D:\a\llama-cpp-python-cuBLAS-wheels\llama-cpp-python-cuBLAS-wheels\vendor\llama.cpp\llama.cpp:3493: codepoints_from_utf8(word).size() > 0

GGML_ASSERT: D:\a\llama-cpp-python-cuBLAS-wheels\llama-cpp-python-cuBLAS-wheels\vendor\llama.cpp\llama.cpp:3493: codepoints_from_utf8(word).size() > 0

I've got the same error message.

Sorry for the inconvenience, I have reported the issue: https://github.com/ggerganov/llama.cpp/issues/6132

Sign up or log in to comment