NVIDIA: Nemotron 3 Super
The NVIDIA Nemotron 3 Super is designed for extensive text handling, with a context length of up to one million tokens and support for tools and reasoning capabilities. It can process inputs beyond simple text, making it versatile for complex tasks that require logical analysis and structured data handling. Given its pricing at $0.085 per million input tokens and $0.4 per million output tokens, the Nemotron 3 Super is suitable for projects with moderate to high computational demands but may be less cost-effective for smaller-scale applications. While it currently lacks independent benchmark coverage, its robust feature set positions it well for users who prioritize reasoning tools and large context lengths.
Benchmark results
Independent, published benchmarks. Blended score 39.4 across 3 benchmarks, last refreshed 2026-07-25. How scoring works →
| Benchmark | Measures | Score |
|---|---|---|
| AI Index | broad capability composite | 41.9 |
| AI Index Coding | software engineering tasks | 62.2 |
| AI Index Agentic | multi-step tool-using tasks | 14.4 |
- Model ID
- nvidia/nemotron-3-super-120b-a12b
- Vendor
- nvidia
- Released
- March 2026
- Tokenizer
- Other
- Input Modalities
- text
- Output Modalities
- text
- Max Output
- 16,384 tokens
- Tool Calling
- ✓ supported
- Structured Output
- ✓ supported
- Reasoning Mode
- ✓ supported
- Vision
- text only
- Audio
- no
- Moderated
- no
What it costs in practice
Computed from the current $0.09/M input and $0.40/M output rates. Run your own numbers →
| Job | Tokens | Cost |
|---|---|---|
| Summarize a 50-page report | 30k in / 1.5k out | under $0.01 |
| Classify 1,000 customer emails | 500k in / 50k out | $0.06 |
| A month of a busy support chatbot | 5M in / 2M out | $1.23 |
Price & spec history
Tracked daily by PicksByModel since 2026-07-17.
| Date | Input /M | Output /M | Context |
|---|---|---|---|
| 2026-07-25 | $0.09 | $0.40 | 1,000,000 |
| 2026-07-21 | $0.08 | $0.45 | 1,000,000 |
| 2026-07-20 | $0.09 | $0.40 | 1,000,000 |
| 2026-07-17 | $0.21 | $0.46 | 1,000,000 |
Category rankings
Where NVIDIA: Nemotron 3 Super places across the 5 categories it ranks in. How we rank →
| # | Category | Score |
|---|---|---|
| #13 | Code CompletionCode · of 25 ranked | 132 |
| #18 | Cheap Bulk InferenceCost · of 25 ranked | 137 |
| #20 | Self-Hosted / LocalCost · of 25 ranked | 117 |
| #23 | Bulk Data LabelingData · of 25 ranked | 132 |
| #25 | Dataset AnnotationResearch · of 25 ranked | 140 |
Similar models
NVIDIA: Nemotron 3 Nano 30B A3B
NVIDIA: Nemotron 3 Nano Omni (free)
NVIDIA: Nemotron 3 Ultra
NVIDIA: Nemotron 3 Super (free)
NVIDIA: Nemotron Nano 12B 2 VL (free)
NVIDIA: Nemotron 3 Ultra (free)
Quick answers
- How much does NVIDIA: Nemotron 3 Super cost?
- $0.09 per million input tokens and $0.40 per million output tokens.
- What is NVIDIA: Nemotron 3 Super's context window?
- 1,000,000 tokens, roughly 1,500 pages of text in a single request.
- Does NVIDIA: Nemotron 3 Super support tool calling?
- It supports tool calling, structured output, a reasoning mode.
- Can NVIDIA: Nemotron 3 Super process images?
- No, it is text-only on the input side.