metadata
base_model: TheDrummer/Magidonia-24B-v4.3
base_model_relation: quantized
quantized_by: ArtusDev
tags:
- exl3
ArtusDev/TheDrummer_Magidonia-24B-v4.3-EXL3
EXL3 quants of TheDrummer/Magidonia-24B-v4.3 using exllamav3 for quantization.
Quants
| Quant | BPW | Head Bits | Size (GB) |
|---|---|---|---|
| 2.5_H6 | 2.5 | 6 | 8.82 |
| 3.0_H6 | 3.0 | 6 | 10.22 |
| 3.5_H6 | 3.5 | 6 | 11.60 |
| 4.0_H6 | 4.0 | 6 | 12.99 |
| 4.5_H6 | 4.5 | 6 | 14.38 |
| 5.0_H6 | 5.0 | 6 | 15.77 |
| 6.0_H6 | 6.0 | 6 | 18.55 |
| 8.0_H8 | 8.0 | 8 | 24.28 |
How to Download and Use Quants
You can download quants by targeting specific size using the Hugging Face CLI.
Click for download commands
1. Install huggingface-cli:
pip install -U "huggingface_hub[cli]"huggingface-cli download ArtusDev/TheDrummer_Magidonia-24B-v4.3-EXL3 --revision "5.0bpw_H6" --local-dir ./EXL3 quants can be run with any inference client that supports EXL3, such as TabbyAPI. Refer to documentation for set up instructions.