Language modelActive
Nemotron 3 Super
NVIDIA
- Released
- March 11, 2026
- Data date
- September 8, 2026
- Speed
- 99.2 tokens/sOutput throughput
- Price
- Input · Output / 1M tokens
- Input
- Text
- Output
- Text
Nemotron 3 Super brings LatentMoE, Mamba-2, attention, and multi-token prediction together in an open text checkpoint. It has 120 billion total parameters, with 12 billion active per token. The checkpoint supports seven languages and belongs to the middle tier of the Nemotron line.
Nemotron 3 Super provides a 1M-token context window. NVIDIA publishes the weights and training recipes.
Specifications and access
| Specification | Value and source |
|---|---|
| Weights | Open weightsSource |
| Access | API, LocalSource |
| Parameters | 120BSource |
| Active parameters | 12BSource |
| Context window | 1,000,000 tokensSource |
| Release | 2026-03-11Source |
| Parameters | 120 BSource |
| Active parameters | 12 BSource |
| License | NVIDIA Nemotron Open Model LicenseSource |
| Architecture | LatentMoE, Mamba-2 + MoE + Attention hybrid with MTPSource |
| Languages | English, French, German, Italian, Japanese, Spanish, ChineseSource |
| Input | TextSource |
| Output | TextSource |
Page 1 of 2
Measurements without matching peer values
These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.
AIME25 (Show measurement, test conditions, and source)
- Source value
- 90.21
- Score
- 90.21
- Metric
- percent
- Unit
- %
- Category
- reasoning
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- NVIDIA model card
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. no tools
AIME25 (Show measurement, test conditions, and source)
- Source value
- 94.73
- Score
- 94.73
- Metric
- percent
- Unit
- %
- Category
- reasoning
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- NVIDIA model card
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. with tools
- Context
- tool-assisted row
GPQA (Show measurement, test conditions, and source)
- Source value
- 79.23
- Score
- 79.23
- Metric
- percent
- Unit
- %
- Category
- reasoning
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- NVIDIA model card
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. no tools
MMLU-Pro (Show measurement, test conditions, and source)
- Source value
- 83.73
- Score
- 83.73
- Metric
- percent
- Unit
- %
- Category
- knowledge
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- NVIDIA model card
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. official eval; no tools unless noted
RULER @ 1M (Show measurement, test conditions, and source)
- Source value
- 91.75
- Score
- 91.75
- Metric
- percent
- Unit
- %
- Category
- long-context
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- NVIDIA model card
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. 1M context
SWE-Bench (OpenHands) (Show measurement, test conditions, and source)
- Source value
- 60.47
- Score
- 60.47
- Metric
- percent
- Unit
- %
- Category
- coding
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- NVIDIA model card
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Vendor-reported result. OpenHands
Artificial Analysis Intelligence Index (Show measurement, test conditions, and source)
- Source value
- 13.5508075290979
- Score
- 13.5508075290979
- Metric
- points
- Unit
- points
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Independent evaluation. unknown
AA-Briefcase Elo (Show measurement, test conditions, and source)
- Source value
- 0
- Score
- 0
- Metric
- elo
- Unit
- elo
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Independent evaluation. unknown
AA-Omniscience Index (Show measurement, test conditions, and source)
- Source value
- -41.5
- Score
- -41.5
- Metric
- points
- Unit
- points
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Independent evaluation. unknown
Output speed (Show measurement, test conditions, and source)
- Source value
- 99.1674413199085
- Score
- 99.1674413199085
- Metric
- tokens/s
- Unit
- tokens/s
- Direction
- Higher is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Independent evaluation. unknown
- Context
- Artificial Analysis performance timing; output speed is decoded after the first streaming chunk. Provider basis: serving-provider median.
Time per Intelligence Index Task (Show measurement, test conditions, and source)
- Source value
- 14.387722601839332
- Score
- 14.387722601839332
- Metric
- minutes
- Unit
- minutes
- Direction
- Lower is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Independent evaluation. unknown
- Context
- Artificial Analysis performance timing; output speed is decoded after the first streaming chunk. Provider basis: serving-provider median.
Latency reasoning time (Show measurement, test conditions, and source)
- Source value
- 20.167909682655964
- Score
- 20.167909682655964
- Metric
- seconds
- Unit
- seconds
- Direction
- Lower is better
- Source type
- independent-evaluator
- Evaluator
- Artificial Analysis
- Status
- active
- Retrieved at
- 2026-09-08
- Methodology
- Independent evaluation. unknown
- Context
- Artificial Analysis performance timing; output speed is decoded after the first streaming chunk. Provider basis: serving-provider median.
Page 1 of 2
Popularity
46.5 / 100
Sources and data date
| Type | Evidence and data date |
|---|---|
| Speed | Median across serving providers. Output speed after the first streaming chunk, not total response time. Retrieved September 8, 2026. |
| Price | Base API prices without caching or batch discounts. Higher context tiers and other rates are listed under Costs. |
| Input | Documented modalities. Retrieved September 8, 2026. |
| Output | Documented modalities. Retrieved September 8, 2026. |
| Research status | Research date September 8, 2026. 2 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication. |
| Not found in the inspected sources | knowledge-cutoff Original wording No exact value located in the checked primary model card/API documentation. |
| Additional source | Nemotron 3 Super technical report · Editorial description reviewed October 2, 2026 |
| Additional source | OpenRouter: Popularity index (derived), CC BY 4.0 |
| Additional source | Vercel AI Gateway: Popularity index (derived), CC BY 4.0 |
| Additional source | Creative Commons Attribution 4.0 |