---
base_model: nvidia/OpenReasoning-Nemotron-7B
---

[EXL3](https://github.com/turboderp-org/exllamav3) quantization of [OpenReasoning-Nemotron-7B](https://huggingface.co/nvidia/OpenReasoning-Nemotron-7B), 8 bits per weight, including output layers.