--- base_model: nvidia/Qwen-3-Nemotron-32B-RLBFF base_model_relation: quantized quantized_by: ArtusDev license: other license_name: nvidia-open-model-license license_link: >- https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/ inference: false fine-tuning: false language: - en tags: - nvidia - qwen3 - exl3 datasets: - nvidia/HelpSteer3 library_name: transformers ---
EXL3 quants of nvidia/Qwen-3-Nemotron-32B-RLBFF using exllamav3 for quantization.
| Quant | BPW | Head Bits | Size (GB) |
|---|---|---|---|
| 2.5_H6 | 2.5 | 6 | 11.93 |
| 3.0_H6 | 3.0 | 6 | 13.88 |
| 3.5_H6 | 3.5 | 6 | 15.83 |
| 4.0_H6 | 4.0 | 6 | 17.78 |
| 4.5_H6 | 4.5 | 6 | 19.73 |
| 5.0_H6 | 5.0 | 6 | 21.68 |
| 6.0_H6 | 6.0 | 6 | 25.58 |
| 8.0_H8 | 8.0 | 8 | 33.57 |
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/nvidia_Qwen-3-Nemotron-32B-RLBFF-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