README.md

metadata

license: other
tags:
  - moe
  - DPO
  - RL-TUNED
  - TensorBlock
  - GGUF
base_model: yunconglong/MoE_13B_DPO

Feedback and support: TensorBlock's Twitter/X, Telegram Group and Discord server

yunconglong/MoE_13B_DPO - GGUF

This repo contains GGUF format model files for yunconglong/MoE_13B_DPO.

The files were quantized using machines provided by TensorBlock, and they are compatible with llama.cpp as of commit b4011.

Prompt template

Model file specification

Filename	Quant type	File Size	Description
MoE_13B_DPO-Q2_K.gguf	Q2_K	4.434 GB	smallest, significant quality loss - not recommended for most purposes
MoE_13B_DPO-Q3_K_S.gguf	Q3_K_S	5.204 GB	very small, high quality loss
MoE_13B_DPO-Q3_K_M.gguf	Q3_K_M	5.780 GB	very small, high quality loss
MoE_13B_DPO-Q3_K_L.gguf	Q3_K_L	6.268 GB	small, substantial quality loss
MoE_13B_DPO-Q4_0.gguf	Q4_0	6.781 GB	legacy; small, very high quality loss - prefer using Q3_K_M
MoE_13B_DPO-Q4_K_S.gguf	Q4_K_S	6.837 GB	small, greater quality loss
MoE_13B_DPO-Q4_K_M.gguf	Q4_K_M	7.248 GB	medium, balanced quality - recommended
MoE_13B_DPO-Q5_0.gguf	Q5_0	8.265 GB	legacy; medium, balanced quality - prefer using Q4_K_M
MoE_13B_DPO-Q5_K_S.gguf	Q5_K_S	8.265 GB	large, low quality loss - recommended
MoE_13B_DPO-Q5_K_M.gguf	Q5_K_M	8.506 GB	large, very low quality loss - recommended
MoE_13B_DPO-Q6_K.gguf	Q6_K	9.842 GB	very large, extremely low quality loss
MoE_13B_DPO-Q8_0.gguf	Q8_0	12.746 GB	very large, extremely low quality loss - not recommended

Downloading instruction

Command line

Firstly, install Huggingface Client

pip install -U "huggingface_hub[cli]"

Then, downoad the individual model file the a local directory

huggingface-cli download tensorblock/MoE_13B_DPO-GGUF --include "MoE_13B_DPO-Q2_K.gguf" --local-dir MY_LOCAL_DIR

If you wanna download multiple model files with a pattern (e.g., *Q4_K*gguf), you can try:

huggingface-cli download tensorblock/MoE_13B_DPO-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'