NVIDIA: Nemotron 3 Super
Analysis Summary
NVIDIA: Nemotron 3 Super sits in the Specialist tier on our leaderboard, ranked #67 of 525 published models on overall intelligence. At $0.090 input and $0.450 output per 1M tokens, it is among the most expensive on the market. It offers a generous context window for extended reasoning and code review and supports tool use and function calling.
Editorial notes
NVIDIA's Nemotron 3 Super offers a solid balance of reasoning and instruction-following capability at very competitive pricing ($0.09/$0.45 per million tokens) with a large 262K context window — a cost-effective choice for businesses running high-volume content or tool-use workflows.
Assessed April 23, 2026
Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?
Performance Profile
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..
Capabilities
Performance Indices
Source: Artificial Analysis
Benchmark Scores
Intelligence
Technical
Content
Benchmark data from Artificial Analysis and Hugging Face
How does NVIDIA: Nemotron 3 Super stack up?
Compare side-by-side with other specialist models.
Model Information
| OpenRouter ID |
nvidia/nemotron-3-super-120b-a12b
|
| Provider | nvidia |
| Release Date | March 11, 2026 |
| Context Length | 262,144 tokens |
| Status | Active |
Pricing
| Token Type | Cost per 1M tokens | Cost per 1K tokens |
|---|---|---|
| Input | $0.09 | $0.000090 |
| Output | $0.45 | $0.000450 |
Live Performance
Live endpoint metrics — refreshed every 30 minutes.
