Skip to content
RankShark

Nemotron 3 Super

NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accurac

ModelOpen sourceby NVIDIAsince 2026
36.7

SharkScore

#177 in Models

Why it scores 36.7

Nemotron 3 Super has a SharkScore of 36.7/100, ranked #177 of 608 in Models (updated 2026-11-19).

Each part compares Nemotron 3 Super with other listings in Models. Evidence covers 87% of the score's weight.

Adoption
7 · 30%
User reviews
no data · 25%
Maintenance
42 · 20%
Trust & transparency
80 · 15%
Momentum
no data · 10%

Paying never changes this score. How ranking works · method v2026-11-18.2

RankShark review of Nemotron 3 Super

Automated evaluation by RankShark · 18 Nov 2026 · method v1.0 · How we review

We were not paid for this review.

As of 2026-11-18, Nemotron 3 Super is an open-weight reasoning model; the OpenRouter listing shows a 262,144-token context and token-based pricing. [E1, E2; MEASURED]

NVIDIA’s 120B-parameter model, with 12B active parameters, is designed for reasoning, chat, tool use, and agent applications. The Hugging Face page describes its hybrid LatentMoE architecture; OpenRouter lists text input and output. [E1, E2]

Best for

  • Developers building agent workflows that use tool calling. · model card
  • Teams working with long-context reasoning or retrieval-augmented generation. · model card
  • High-volume applications such as IT ticket automation. · model card

Key features

  • Hybrid model architecture. The vendor describes a LatentMoE architecture combining Mamba-2, mixture-of-experts layers, attention, and Multi-Token Prediction. · model card
  • Configurable reasoning. The vendor says reasoning can be enabled or disabled through the chat template. OpenRouter lists reasoning efforts of medium, low, and none. · model card, OpenRouter
  • Tool use. The vendor says the model is intended for tool use and agent workflows; OpenRouter lists support for tools and tool_choice parameters. · model card, OpenRouter
  • Multilingual support. The vendor lists English, French, German, Italian, Japanese, Spanish, and Chinese. · model card
  • OpenRouter context and output limits. The OpenRouter listing reports a 262,144-token context window and up to 16,384 completion tokens. · OpenRouter

Pros

  • The vendor identifies agent workflows, tool use, RAG, and long-context reasoning as intended uses. · model card
  • OpenRouter lists several API endpoints, including chat completions, completions, messages, and responses. · OpenRouter
  • The Hugging Face page provides model weights and describes the model as ready for commercial use under the NVIDIA Nemotron Open Model License. · model card

Cons

  • For self-hosting, the vendor lists a minimum GPU requirement of 2× H100-80GB. · model card
  • The Hugging Face page says the model supports up to 1M tokens, while the OpenRouter listing reports a 262,144-token context window. · model card, OpenRouter
  • No free plan or free trial stated on the pages we checked.
  • Whether customer data is used for model training is not stated on the pages we checked.

Pricing

PlanPriceLimits
OpenRouter API$0.075 per million input tokens; $0.40 per million output tokens262,144-token context window; up to 16,384 completion tokens

Checked 18 Nov 2026

Free plan
Not stated (checked 18 Nov 2026)
Free trial
Not stated (checked 18 Nov 2026)
Trains on customer data
Not stated (checked 18 Nov 2026)
Data retention
Not stated (checked 18 Nov 2026)
Data regions
Not stated (checked 18 Nov 2026)
Data processing agreement
Not stated (checked 18 Nov 2026)
AI models used
Not stated (checked 18 Nov 2026)
API
OpenRouter lists chat completions, completions, messages, and responses endpoints. · OpenRouter
SDKs
Not stated (checked 18 Nov 2026)
Webhooks
Not stated (checked 18 Nov 2026)
Platforms
The model weights are listed on Hugging Face, and an API listing is available on OpenRouter. · model card, OpenRouter
Security claims
No security certification stated (checked 18 Nov 2026)

What we measured

Context window (OpenRouter listing) · now
262,144 tokens
Price per million tokens (OpenRouter listing) · now
$0.075 in, $0.4 out
Package downloads · 30 days
28,697
How we tested · sources
  • Hourly uptime and response-time checks of the product's URL.
  • Repository, release and download data from GitHub, npm, PyPI and Hugging Face where they exist.
  • We did not sign up, pay, or use the product as a customer.
  • 18 Nov 2026: Plans or prices changed on the pricing page.
  • 18 Nov 2026: Plans or prices changed on the pricing page.

Something wrong? Report an error with a public link that shows it; corrections re-run the review.

Top alternatives to Nemotron 3 Super

See all
  • 67.1

    Qwen3.8 27B is an open-weight dense vision-language model from Qwen.

    ModelOpen source
  • 65.5

    Qwen3.6-35B-A3B is an open-weight multimodal model from Alibaba Cloud with 35 billion total parameters and 3 billion active parameters per t

    ModelOpen source
  • 64.7

    Qwen3.8 Flash is a multimodal reasoning model from Alibaba.

    ModelOpen source
  • 64.3

    Qwen3.6 27B is a dense 27-billion-parameter language model from the Qwen Team at Alibaba, released in April 2026.

    ModelOpen source