Nemotron 3 Ultra
Released on June 4, 2026
Changes from previous version
Moves from the compact multimodal Nemotron 3 Nano Omni to a 550B-total, 55B-active text reasoning engine focused on the hardest long-running agent and planning workloads.
Release Summary
NVIDIA's open 550B-total, 55B-active reasoning engine is the largest Nemotron 3 model and is designed for long-running agents, complex planning, multi-step tool use, coding, and enterprise workflows with up to 1M tokens of context.
Timeline
June 4, 2026
Nemotron 3 Ultra released
NVIDIA releases its largest open Nemotron 3 reasoning model for long-running agents.