Get Access to NVIDIA Nemotron-3 8B Models
The NVIDIA Nemotron-3 8B family of models is optimized for building production-ready generative AI applications for the enterprise. The Nemotron-3 8B base, chat, and aligned models can be customized with various techniques, including domain-adapted pretraining, SFT, LoRA, SteerLM, and p-tuning using the NVIDIA NeMo™ framework.
| Model | Variant | Key Benefits |
|---|---|---|
| Base | Nemotron-3-8B-Base | Enables customization, including parameter-efficient fine-tuning and continuous pre-training for domain-adapted LLMs |
| Chat | Nemotron-3-8B-Chat-SFT | A building block for instruction tuning custom models or user-defined alignment, such as RLHF or SteerLM models |
| Nemotron-3-8B-Chat-RLHF | Best out-of-the-box chat model performance | |
| Nemotron-3-8B-Chat-SteerLM | Best out-of-the-box chat model with flexible alignment at inference time | |
| Question-and- Answer | Nemotron-3-8B-QA | Q&A LLMs customized on knowledge bases |
To download these models, you must be a member of the NVIDIA Developer Program and logged in with your organization's email address.