GGUF
TensorBlock
GGUF
Viking-33B-GGUF / README.md
morriszms's picture
Update README.md
10e3a5a verified
metadata
license: apache-2.0
datasets:
  - cerebras/SlimPajama-627B
  - bigcode/starcoderdata
  - mc4
language:
  - fi
  - en
  - da
  - sv
  - 'no'
  - nn
  - is
tags:
  - TensorBlock
  - GGUF
base_model: LumiOpen/Viking-33B
TensorBlock

Website Twitter Discord GitHub Telegram

LumiOpen/Viking-33B - GGUF

This repo contains GGUF format model files for LumiOpen/Viking-33B.

The files were quantized using machines provided by TensorBlock, and they are compatible with llama.cpp as of commit b4242.

Our projects

Forge
Forge Project
An OpenAI-compatible multi-provider routing layer.
πŸš€ Try it now! πŸš€
Awesome MCP Servers TensorBlock Studio
MCP Servers Studio
A comprehensive collection of Model Context Protocol (MCP) servers. A lightweight, open, and extensible multi-LLM interaction studio.
πŸ‘€ See what we built πŸ‘€ πŸ‘€ See what we built πŸ‘€
## Prompt template

Model file specification

Filename Quant type File Size Description
Viking-33B-Q2_K.gguf Q2_K 12.561 GB smallest, significant quality loss - not recommended for most purposes
Viking-33B-Q3_K_S.gguf Q3_K_S 14.606 GB very small, high quality loss
Viking-33B-Q3_K_M.gguf Q3_K_M 16.191 GB very small, high quality loss
Viking-33B-Q3_K_L.gguf Q3_K_L 17.573 GB small, substantial quality loss
Viking-33B-Q4_0.gguf Q4_0 18.879 GB legacy; small, very high quality loss - prefer using Q3_K_M
Viking-33B-Q4_K_S.gguf Q4_K_S 19.012 GB small, greater quality loss
Viking-33B-Q4_K_M.gguf Q4_K_M 19.992 GB medium, balanced quality - recommended
Viking-33B-Q5_0.gguf Q5_0 22.902 GB legacy; medium, balanced quality - prefer using Q4_K_M
Viking-33B-Q5_K_S.gguf Q5_K_S 22.902 GB large, low quality loss - recommended
Viking-33B-Q5_K_M.gguf Q5_K_M 23.475 GB large, very low quality loss - recommended
Viking-33B-Q6_K.gguf Q6_K 27.175 GB very large, extremely low quality loss
Viking-33B-Q8_0.gguf Q8_0 35.196 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/Viking-33B-GGUF --include "Viking-33B-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/Viking-33B-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'