Hugging Face's logo

tensorblock
/
deepseek-math-7b-rl-GGUF

GGUF
TensorBlock
GGUF
Inference Endpoints
conversational
Model card Files Community
deepseek-math-7b-rl-GGUF / README.md
morriszms's picture
morriszms
Update README.md
b87f342 verified
preview code
|
raw
history blame contribute delete
4.92 kB
metadata
license: other
license_name: deepseek
license_link: https://github.com/deepseek-ai/DeepSeek-Math/blob/main/LICENSE-MODEL
tags:
  - TensorBlock
  - GGUF
base_model: deepseek-ai/deepseek-math-7b-rl
TensorBlock

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

deepseek-ai/deepseek-math-7b-rl - GGUF

This repo contains GGUF format model files for deepseek-ai/deepseek-math-7b-rl.

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

Prompt template 

<｜begin▁of▁sentence｜>{system_prompt}

User: {prompt}

Assistant:

Model file specification

Filename Quant type File Size Description
deepseek-math-7b-rl-Q2_K.gguf Q2_K 2.532 GB smallest, significant quality loss - not recommended for most purposes
deepseek-math-7b-rl-Q3_K_S.gguf Q3_K_S 2.923 GB very small, high quality loss
deepseek-math-7b-rl-Q3_K_M.gguf Q3_K_M 3.223 GB very small, high quality loss
deepseek-math-7b-rl-Q3_K_L.gguf Q3_K_L 3.489 GB small, substantial quality loss
deepseek-math-7b-rl-Q4_0.gguf Q4_0 3.725 GB legacy; small, very high quality loss - prefer using Q3_K_M
deepseek-math-7b-rl-Q4_K_S.gguf Q4_K_S 3.749 GB small, greater quality loss
deepseek-math-7b-rl-Q4_K_M.gguf Q4_K_M 3.933 GB medium, balanced quality - recommended
deepseek-math-7b-rl-Q5_0.gguf Q5_0 4.481 GB legacy; medium, balanced quality - prefer using Q4_K_M
deepseek-math-7b-rl-Q5_K_S.gguf Q5_K_S 4.481 GB large, low quality loss - recommended
deepseek-math-7b-rl-Q5_K_M.gguf Q5_K_M 4.588 GB large, very low quality loss - recommended
deepseek-math-7b-rl-Q6_K.gguf Q6_K 5.284 GB very large, extremely low quality loss
deepseek-math-7b-rl-Q8_0.gguf Q8_0 6.842 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-math-7b-rl-GGUF --include "deepseek-math-7b-rl-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-math-7b-rl-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'