mirror of
https://github.com/ggml-org/llama.cpp.git
synced 2026-11-09 13:24:42 +01:00
* gguf_convert_endian.py: skip MXFP4 data * Use gguf.constants.GGML_QUANT_SIZES to determine block sizes
* gguf_convert_endian.py: skip MXFP4 data * Use gguf.constants.GGML_QUANT_SIZES to determine block sizes