Model Releases

NVIDIA releases open-source Nemotron 3.5 Super VL and Bolt edge models

Original title:Open From Data Center to Desktop: NVIDIA Expands Nemotron Family With Highest Efficiency Super VL and Bolt Models

AI summary

NVIDIA released the open-source 120B-parameter multimodal model Nemotron 3.5 Super VL, aimed at high-frequency agent tasks; the company says its accuracy is comparable to its largest reasoning model at about one-fifth the parameters, and throughput is up to 5x faster than the next-best comparable model. It also launched edge models Bolt 74B-A9B 24GB (native 2-bit quantization, runs locally on RTX/DGX Spark) and Bolt E4B 4GB (available October 23).

Key Takeaways

  • NVIDIA released the open-source 120B-parameter multimodal model Nemotron 3.
  • 5 Super VL,aimed at high-frequency agent tasks
  • the company says its accuracy is comparable to its largest reasoning model at about one-fifth the…
  • It also launched edge models Bolt 74B-A9B 24GB (native 2-bit quantization,runs locally on RTX/DGX…

Quick Facts

Published
Updated
Source
NVIDIA Blog(RSS)
Category
Model Releases
Topic
Model ReleasesOpen-Source EcosystemMultimodal
AI score
71
Read the original Read the full article at the original source
Why it matters

原文给出三款开放模型各自针对的部署层级和量化、训练数据规模,读者可据此判断哪一档适合自家智能体场景。

Topics

Related entities

Source & copyright

This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.

Original article · AIHOT entry

How to Cite

When citing this article, keep the title, date, and permalink, and credit Orange News as the source.

Orange News: NVIDIA releases open-source Nemotron 3.5 Super VL and Bolt edge models. October 15, 2026. https://new.nat6.net/en/news/l9jgzg24ei3a7m8xu1xfhy9n7/

View all →