metadata
base_model: internlm/JanusCoder-14B
base_model_relation: quantized
quantized_by: ArtusDev
license: apache-2.0
library_name: transformers
tags:
- exl3
ArtusDev/internlm_JanusCoder-14B-EXL3
EXL3 quants of internlm/JanusCoder-14B using exllamav3 for quantization.
Quants
| Quant | BPW | Head Bits | Size (GB) |
|---|---|---|---|
| 2.5_H6 | 2.5 | 6 | 6.29 |
| 3.0_H6 | 3.0 | 6 | 7.12 |
| 3.5_H6 | 3.5 | 6 | 7.94 |
| 4.0_H6 | 4.0 | 6 | 8.77 |
| 4.5_H6 | 4.5 | 6 | 9.60 |
| 5.0_H6 | 5.0 | 6 | 10.42 |
| 6.0_H6 | 6.0 | 6 | 12.07 |
| 8.0_H8 | 8.0 | 8 | 15.57 |
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/internlm_JanusCoder-14B-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.
Acknowledgements
Made possible with cloud compute from lium.io