Model CoverageLarge Language ModelsNVIDIAMinitron-8B-Base

Minitron-8B-Base

View as Markdown

NVIDIA Nemotron and Minitron are NVIDIA’s family of language models. Minitron models are produced by pruning and distilling larger Llama/Nemotron models into compact, high-performance checkpoints.

Use this page as a checkpoint and architecture reference. Set up NeMo AutoModel with the latest container or follow the installation instructions.

Model Reference

Model Architecture

PropertyValue
TaskText Generation
ArchitectureNemotronForCausalLM
Parameters8B
Hugging Face Organizationnvidia

Available Models

ModelHF ID
Minitron 8B Basenvidia/Minitron-8B-Base