NVIDIA (NVDA) Releases New Nemotron 3 Super Model to Power Advanced Agentic AI

FomoAnxiety · 2026-03-11T17:22:28+00:00

NVIDIA has launched Nemotron 3 Super, a high-accuracy AI model designed for complex multi-agent systems and workflows, featuring 120 billion parameters and enhanced performance metrics. It is already being adopted across various industries for improved accuracy.

FomoAnxiety

2026-03-11 17:22:28

Abstract generation in progress

NVIDIA NVDA +0.38% ▲ has released Nemotron 3 Super as part of its growing family of open AI models. This high-accuracy reasoning model is specifically designed to power complex multi-agent systems and enterprise AI workflows.

Claim 70% Off TipRanks Premium

Unlock hedge fund-level data and powerful investing tools for smarter, sharper decisions
Stay ahead of the market with the latest news and analysis and maximize your portfolio’s potential

The model features a total of 120 billion parameters, with 12 billion active parameters per token. Nemotron 3 Super is built especially for autonomous agents that need long‑horizon reasoning, tool use, and the ability to keep track of context across many steps.

NVIDIA said it delivers up to 5× higher throughput, up to 2× better accuracy, and a massive 1‑million‑token context window, which helps prevent “goal drift” during complex, multi‑step tasks.

The model uses a hybrid MoE architecture that blends Transformer and Mamba components, enabling high reasoning performance while keeping compute costs manageable.

Importantly, the model is released with open weights under a permissive license, allowing developers to inspect, customize, and deploy it on their own infrastructure.

Adoption Across Industries

Several companies have already begun integrating Nemotron 3 Super. Perplexity is offering it as part of its search and model‑orchestration stack. Also, CodeRabbit, Factory, and Greptile are using it to boost accuracy in software‑development agents.

At the same time, Edison Scientific and Lila Sciences are applying it to deep literature search and molecular analysis.

Moreover, enterprise platforms such as Amdocs DOX -1.49% ▼ , Palantir PLTR +0.22% ▲ , Cadence CDNS -1.19% ▼ , Dassault DASTY -0.19% ▼ , and Siemens SIEGY -1.00% ▼ are customizing it for telecom, cybersecurity, semiconductor design, and manufacturing workflows.

Is NVDA a Strong Buy?

Overall, Wall Street has a Strong Buy consensus rating on Nvidia stock based on 38 Buy and one Hold recommendations. The average NVDA stock price target of $273.61 indicates 47.43% upside potential.

Disclaimer & DisclosureReport an Issue

This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.