Nemotron 3 Ultra: Open Moe Hybrid Mamba-Transformer for Agentic Reasoning [pdf]

victormustar 23 points 2 comments June 04, 2026
research.nvidia.com · View on Hacker News

Discussion Highlights (2 comments)

throwa356262

Is this the one from Jensens Computex presentation the other day? It is significantly bigger than Qwen for the same level of intelligence, but I think the key strength was inference speed.

2001zhaozhao

This model seems like a really big deal. Is this the biggest Western open-source AI model in the world (beating out Llama3 405B)?

Semantic search powered by Rivestack pgvector
10,324 stories · 97,050 chunks indexed