Misaka27260's picture
Update README.md
d50291d verified
metadata
license: apache-2.0
language:
  - zh
  - en
base_model:
  - kxdw2580/Qwen2.5-3B-Instruct-Catgirl-v2
pipeline_tag: text-generation
tags:
  - qwen2
  - conversational

Quantized and converted gguf file from https://huggingface.co/kxdw2580/Qwen2.5-3B-Instruct-Catgirl-v2

Obviously the parameter number is rather small, thus it is not a powerful enough model, but its reply is interesting. In a sense it's not a useless model.

Since parameter number is small, FP16 precision is recomended.