tensorblock/Mixtral_7Bx5_MoE_30B-GGUF

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

cloudyu/Mixtral_7Bx5_MoE_30B - GGUF

This repo contains GGUF format model files for cloudyu/Mixtral_7Bx5_MoE_30B.

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

Filename	Quant type	File Size	Description
Mixtral_7Bx5_MoE_30B-Q2_K.gguf	Q2_K	10.884 GB	smallest, significant quality loss - not recommended for most purposes
Mixtral_7Bx5_MoE_30B-Q3_K_S.gguf	Q3_K_S	12.856 GB	very small, high quality loss
Mixtral_7Bx5_MoE_30B-Q3_K_M.gguf	Q3_K_M	14.267 GB	very small, high quality loss
Mixtral_7Bx5_MoE_30B-Q3_K_L.gguf	Q3_K_L	15.451 GB	small, substantial quality loss
Mixtral_7Bx5_MoE_30B-Q4_0.gguf	Q4_0	16.795 GB	legacy; small, very high quality loss - prefer using Q3_K_M
Mixtral_7Bx5_MoE_30B-Q4_K_S.gguf	Q4_K_S	16.944 GB	small, greater quality loss
Mixtral_7Bx5_MoE_30B-Q4_K_M.gguf	Q4_K_M	18.024 GB	medium, balanced quality - recommended
Mixtral_7Bx5_MoE_30B-Q5_0.gguf	Q5_0	20.502 GB	legacy; medium, balanced quality - prefer using Q4_K_M
Mixtral_7Bx5_MoE_30B-Q5_K_S.gguf	Q5_K_S	20.502 GB	large, low quality loss - recommended
Mixtral_7Bx5_MoE_30B-Q5_K_M.gguf	Q5_K_M	21.135 GB	large, very low quality loss - recommended
Mixtral_7Bx5_MoE_30B-Q6_K.gguf	Q6_K	24.442 GB	very large, extremely low quality loss
Mixtral_7Bx5_MoE_30B-Q8_0.gguf	Q8_0	31.656 GB	very large, extremely low quality loss - not recommended

Firstly, install Huggingface Client

pip install -U "huggingface_hub[cli]"

Then, downoad the individual model file the a local directory

huggingface-cli download tensorblock/Mixtral_7Bx5_MoE_30B-GGUF --include "Mixtral_7Bx5_MoE_30B-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/Mixtral_7Bx5_MoE_30B-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'