NVIDIA: Nemotron 3 Super
Analysis Summary
NVIDIA Nemotron 3 Super is designed for agentic workflows, pairing a million-token context with tool use and function calling. Its coding capability is strong and its instruction-following result is useful for structured operations, making it suitable for assistants that retrieve information, call services, and maintain substantial working context.
The model fits document processing, internal operations, coding support, and controlled automation with explicit tool boundaries. General reasoning is less capable than premium flagship models, and tool reliability is useful rather than exceptional, so high-impact actions should retain validation and approval steps. Its low listed input and output prices create a favourable volume profile. Use it as an economical agent layer, not as the sole model for complex strategy or nuanced client writing.
Assessed August 9, 2026
Editorial notes
Nemotron 3 Super is a strong value-oriented agent model with a million-token context, tool use, function calling, and capable coding performance. General reasoning is more limited, so it fits structured automation better than open-ended strategic work.
Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?
DFO Verdict
Nemotron 3 Super is a strong value-oriented agent model with a million-token context, tool use, function calling, and capable coding performance. General reasoning is more limited, so it fits structured automation better than open-ended strategic work.
Benchmark scores
Magenta = intelligence · Ink = technical/agentic · Cyan = content & long-context · Grey = community benchmarks. Data: Artificial Analysis, Hugging Face.
25.4 Intelligence Index·37.7 Coding Index·48.3 Agentic Index
How NVIDIA: Nemotron 3 Super compares
NVIDIA: Nemotron 3 Super ranks #147 of 425 AI models we track for overall intelligence, #94 of 198 for coding, #25 of 180 for agentic tasks. Its 1M-token context window is larger than 86% of the models we list. At $0.09 per million input tokens it is cheaper than 71% of comparable models.
Dark bar = input · light bar = output, scaled to the priciest peer.
1M tokens ≈ 8 full-length novels or ~2,500 pages of business documents in a single request.
Strongest on value. The pulled-in intelligence corner is the trade-off, and if the shape matters more than the price, this is your model.
Compare shapes side-by-side →Pricing
| Token Type | Cost per 1M tokens | Cost per 1K tokens |
|---|---|---|
| Input | $0.09 | $0.000085 |
| Output | $0.40 | $0.000400 |
What would NVIDIA: Nemotron 3 Super cost your business?
Pick the job that looks most like yours, then fine-tune with the sliders. Estimates update live.
A website chatbot handling around 100 customer conversations a day, a few short messages each.
Full calculator with 704 models → Price Calculator
These numbers get smaller with the right architecture.
We route routine calls to cheap models and save NVIDIA: Nemotron 3 Super for the hard ones. Most clients cut their estimate by 60-80%.
Talk to our teamAbout NVIDIA: Nemotron 3 Super
NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications. Built on a hybrid Mamba-Transformer..
Frequently asked questions about NVIDIA: Nemotron 3 Super
How much does NVIDIA: Nemotron 3 Super cost?
NVIDIA: Nemotron 3 Super costs $0.09 per million input tokens and $0.40 per million output tokens.
What is the context window of NVIDIA: Nemotron 3 Super?
NVIDIA: Nemotron 3 Super has a context window of 1,000,000 tokens (1M).
Is NVIDIA: Nemotron 3 Super good for coding?
On our coding benchmark index, NVIDIA: Nemotron 3 Super ranks #94 of 198 models, placing it in the broader range of the field for code generation and debugging.
What can NVIDIA: Nemotron 3 Super do?
NVIDIA: Nemotron 3 Super supports tool use and function calling.
Who created NVIDIA: Nemotron 3 Super?
NVIDIA: Nemotron 3 Super is developed by NVIDIA and was released on March 11, 2026.