NVIDIA Nemotron 3 Ultra technical report

NVIDIA released a 550B-total, 55B-active hybrid Mamba-attention MoE model with an open post-training stack aimed at agentic workloads

Proud to see Nemotron 3 Ultra out! From Nano → Super → Ultra, we’ve kept pushing the post-training stack: SFT → multi-env RLVR → MOPD with many specialized teachers. The result: fast, token-efficient, open, and strong on agentic tasks. T

Ranked #6 on backlist 2026-06-04 (04 Jun 2026 UTC) · by @soumyesinghal (Soumye Singhal) · view on X ↗

How it ranks: Backlist reads my Twitter/X timeline, scores every tweet for substance with an LLM rubric (not engagement), and publishes the daily top picks with a one-line takeaway. Curated by Surya Dantuluri.