tensorblock/Moonlight-L3-15B-v2-64k-GGUF

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

Darkknight535/Moonlight-L3-15B-v2-64k - GGUF

This repo contains GGUF format model files for Darkknight535/Moonlight-L3-15B-v2-64k.

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

Prompt template

<|begin_of_text|><|start_header_id|>system<|end_header_id|>

{system_prompt}<|eot_id|><|start_header_id|>user<|end_header_id|>

{prompt}<|eot_id|><|start_header_id|>assistant<|end_header_id|>

Model file specification

Filename	Quant type	File Size	Description
Moonlight-L3-15B-v2-64k-Q2_K.gguf	Q2_K	5.352 GB	smallest, significant quality loss - not recommended for most purposes
Moonlight-L3-15B-v2-64k-Q3_K_S.gguf	Q3_K_S	6.207 GB	very small, high quality loss
Moonlight-L3-15B-v2-64k-Q3_K_M.gguf	Q3_K_M	6.866 GB	very small, high quality loss
Moonlight-L3-15B-v2-64k-Q3_K_L.gguf	Q3_K_L	7.431 GB	small, substantial quality loss
Moonlight-L3-15B-v2-64k-Q4_0.gguf	Q4_0	7.998 GB	legacy; small, very high quality loss - prefer using Q3_K_M
Moonlight-L3-15B-v2-64k-Q4_K_S.gguf	Q4_K_S	8.055 GB	small, greater quality loss
Moonlight-L3-15B-v2-64k-Q4_K_M.gguf	Q4_K_M	8.482 GB	medium, balanced quality - recommended
Moonlight-L3-15B-v2-64k-Q5_0.gguf	Q5_0	9.684 GB	legacy; medium, balanced quality - prefer using Q4_K_M
Moonlight-L3-15B-v2-64k-Q5_K_S.gguf	Q5_K_S	9.684 GB	large, low quality loss - recommended
Moonlight-L3-15B-v2-64k-Q5_K_M.gguf	Q5_K_M	9.934 GB	large, very low quality loss - recommended
Moonlight-L3-15B-v2-64k-Q6_K.gguf	Q6_K	11.476 GB	very large, extremely low quality loss
Moonlight-L3-15B-v2-64k-Q8_0.gguf	Q8_0	14.861 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/Moonlight-L3-15B-v2-64k-GGUF --include "Moonlight-L3-15B-v2-64k-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/Moonlight-L3-15B-v2-64k-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'