--- base_model: yamatazen/EsotericSage-12B base_model_relation: quantized quantized_by: ArtusDev tags: - mergekit - merge - nearswap - exl3 language: - en - ja ---
EXL3 quants of yamatazen/EsotericSage-12B using exllamav3 for quantization.
| Quant | BPW | Head Bits | Size (GB) |
|---|---|---|---|
| 2.5_H6 | 2.5 | 6 | 5.28 |
| 3.0_H6 | 3.0 | 6 | 5.97 |
| 3.5_H6 | 3.5 | 6 | 6.65 |
| 4.0_H6 | 4.0 | 6 | 7.33 |
| 4.5_H6 | 4.5 | 6 | 8.01 |
| 5.0_H6 | 5.0 | 6 | 8.69 |
| 6.0_H6 | 6.0 | 6 | 10.06 |
| 8.0_H8 | 8.0 | 8 | 12.95 |
You can download quants by targeting specific size using the Hugging Face CLI.
pip install -U "huggingface_hub[cli]"
2. Download a specific quant:
huggingface-cli download ArtusDev/yamatazen_EsotericSage-12B-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.