NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 is a large language model released by NVIDIA, released under the other license, with a claimed knowledge cutoff of 2026-06-04. Its weights, configuration, and tokenizer files are shortlisted for hashing and on-chain attestation by Weight Registry.
Description
NVIDIA Nemotron 3 Ultra is the largest tier of NVIDIA's Nemotron 3 open-model family, released June 4, 2026, with 550B total parameters and roughly 55B active via Mixture-of-Experts routing on a hybrid Transformer-Mamba architecture. It's built for frontier-level agentic reasoning and orchestration workloads.
Metadata
Hugging Face stats
233,434 downloads on Hugging Face · 350 likes · pipeline: text-generation · library: transformers
Weight Hashes
Not yet hashed -- queued for processing.