Hugging Face's logo

tensorblock
/
Swallow-7b-instruct-hf-GGUF

Text Generation
Transformers
GGUF
English
Japanese
TensorBlock
GGUF
Inference Endpoints
Model card Files Community
Swallow-7b-instruct-hf-GGUF / README.md
morriszms's picture
morriszms
Update README.md
f25fb26 verified
preview code
|
raw
history blame contribute delete
4.97 kB
metadata
language:
  - en
  - ja
library_name: transformers
pipeline_tag: text-generation
license: llama2
model_type: llama
tags:
  - TensorBlock
  - GGUF
base_model: tokyotech-llm/Swallow-7b-instruct-hf
TensorBlock

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

tokyotech-llm/Swallow-7b-instruct-hf - GGUF

This repo contains GGUF format model files for tokyotech-llm/Swallow-7b-instruct-hf.

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
Swallow-7b-instruct-hf-Q2_K.gguf Q2_K 2.408 GB smallest, significant quality loss - not recommended for most purposes
Swallow-7b-instruct-hf-Q3_K_S.gguf Q3_K_S 2.799 GB very small, high quality loss
Swallow-7b-instruct-hf-Q3_K_M.gguf Q3_K_M 3.125 GB very small, high quality loss
Swallow-7b-instruct-hf-Q3_K_L.gguf Q3_K_L 3.404 GB small, substantial quality loss
Swallow-7b-instruct-hf-Q4_0.gguf Q4_0 3.622 GB legacy; small, very high quality loss - prefer using Q3_K_M
Swallow-7b-instruct-hf-Q4_K_S.gguf Q4_K_S 3.651 GB small, greater quality loss
Swallow-7b-instruct-hf-Q4_K_M.gguf Q4_K_M 3.860 GB medium, balanced quality - recommended
Swallow-7b-instruct-hf-Q5_0.gguf Q5_0 4.397 GB legacy; medium, balanced quality - prefer using Q4_K_M
Swallow-7b-instruct-hf-Q5_K_S.gguf Q5_K_S 4.397 GB large, low quality loss - recommended
Swallow-7b-instruct-hf-Q5_K_M.gguf Q5_K_M 4.519 GB large, very low quality loss - recommended
Swallow-7b-instruct-hf-Q6_K.gguf Q6_K 5.220 GB very large, extremely low quality loss
Swallow-7b-instruct-hf-Q8_0.gguf Q8_0 6.760 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/Swallow-7b-instruct-hf-GGUF --include "Swallow-7b-instruct-hf-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/Swallow-7b-instruct-hf-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'