tensorblock/DeepSeek-V2-Lite-Chat-GGUF

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

deepseek-ai/DeepSeek-V2-Lite-Chat - GGUF

This repo contains GGUF format model files for deepseek-ai/DeepSeek-V2-Lite-Chat.

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

Run them on the TensorBlock client using your local machine ↗

Prompt template

<｜begin▁of▁sentence｜>{system_prompt}

User: {prompt}

Assistant:

Model file specification

Filename	Quant type	File Size	Description
DeepSeek-V2-Lite-Chat-Q2_K.gguf	Q2_K	5.989 GB	smallest, significant quality loss - not recommended for most purposes
DeepSeek-V2-Lite-Chat-Q3_K_S.gguf	Q3_K_S	6.973 GB	very small, high quality loss
DeepSeek-V2-Lite-Chat-Q3_K_M.gguf	Q3_K_M	7.568 GB	very small, high quality loss
DeepSeek-V2-Lite-Chat-Q3_K_L.gguf	Q3_K_L	7.878 GB	small, substantial quality loss
DeepSeek-V2-Lite-Chat-Q4_0.gguf	Q4_0	8.294 GB	legacy; small, very high quality loss - prefer using Q3_K_M
DeepSeek-V2-Lite-Chat-Q4_K_S.gguf	Q4_K_S	8.879 GB	small, greater quality loss
DeepSeek-V2-Lite-Chat-Q4_K_M.gguf	Q4_K_M	9.653 GB	medium, balanced quality - recommended
DeepSeek-V2-Lite-Chat-Q5_0.gguf	Q5_0	10.097 GB	legacy; medium, balanced quality - prefer using Q4_K_M
DeepSeek-V2-Lite-Chat-Q5_K_S.gguf	Q5_K_S	10.378 GB	large, low quality loss - recommended
DeepSeek-V2-Lite-Chat-Q5_K_M.gguf	Q5_K_M	11.037 GB	large, very low quality loss - recommended
DeepSeek-V2-Lite-Chat-Q6_K.gguf	Q6_K	13.101 GB	very large, extremely low quality loss
DeepSeek-V2-Lite-Chat-Q8_0.gguf	Q8_0	15.555 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/DeepSeek-V2-Lite-Chat-GGUF --include "DeepSeek-V2-Lite-Chat-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/DeepSeek-V2-Lite-Chat-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'