NVIDIA Drops a Massive Open Bundle: Nemotron Models, 10T Tokens, and an LLM Router Blueprint

NVIDIA Blog·medium signalXBlueskyLinkedInCopy link

On June 22 NVIDIA released open models, datasets, and tools across the Nemotron family — including training data and code for the Llama Embed Nemotron 8B embedding model (on the MMTEB leaderboard), an updated LLM Router blueprint that auto-directs requests to the best model for a job, and the dataset behind Nemotron Speech ASR. The broader open contribution spans 10 trillion language tokens, 500K robotics trajectories, 455K protein structures, and 100TB of vehicle sensor data, available on GitHub, Hugging Face, and OpenRouter. Verified on the NVIDIA blog.