--- base_model: ilgee/Multiclass-Think-RM-8B base_model_relation: quantized quantized_by: ArtusDev license: llama3.1 language: - en tags: - reward-model - RLHF - reasoning - preference-learning - exl3 ---
EXL3 quants of ilgee/Multiclass-Think-RM-8B using exllamav3 for quantization.
| Quant | BPW | Head Bits | Size (GB) |
|---|---|---|---|
| 2.5_H6 | 2.5 | 6 | 3.65 |
| 3.0_H6 | 3.0 | 6 | 4.09 |
| 3.5_H6 | 3.5 | 6 | 4.52 |
| 4.0_H6 | 4.0 | 6 | 4.96 |
| 4.5_H6 | 4.5 | 6 | 5.40 |
| 5.0_H6 | 5.0 | 6 | 5.83 |
| 6.0_H6 | 6.0 | 6 | 6.71 |
| 8.0_H8 | 8.0 | 8 | 8.58 |
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/ilgee_Multiclass-Think-RM-8B-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.
Made possible with cloud compute from lium.io