IQ3 quants of NousResearch/Nous-Hermes-2-Yi-34B
Created using llama.cpp 9e359a4f, with default settings of both convert.py and quantize using the imatrix provided by ikawrakow.
See https://github.com/ggerganov/llama.cpp/pull/5676 for information on the IQ3 quantization.
- Downloads last month
- 17
3-bit
Model tree for patf82/Nous-Hermes-2-Yi-34B-IQ3-imatrix-GGUF
Base model
01-ai/Yi-34B
Quantized
this model