Nemotron 3 Nano
API request failed. The model was not evaluated.
Language modelActive
Nemotron 3 Nano
NVIDIA
- Released
- December 15, 2025
- Data date
- September 8, 2026
- Predecessor
- Nemotron Nano 2 9B
- Speed
- 222.6 tokens/sOutput throughput
- Price
- Input · Output / 1M tokens
- Input
- Text
- Output
- Text
Nemotron 3 Nano's new step beyond Nemotron Nano 2 9B is a sparse mixture-of-experts design. The 30-billion-parameter checkpoint activates 3.5 billion parameters per token and combines Mamba2, Transformer, and attention. Reasoning and an explicit thinking budget make the planning share configurable.
NVIDIA lists up to 1M context tokens, six supported languages, and coding with tool use. The open weights are intended for custom inference. The 3.5 billion active parameters describe the per-token activation, not the checkpoint's total memory requirement.
Specifications and access
| Specification | Value and source |
|---|---|
| Weights | Open weightsSource |
| Access | API, LocalSource |
| Parameters | 30BSource |
| Active parameters | 3.5BSource |
| Context window | 1,000,000 tokensSource |
| Knowledge cutoff | June 25, 2025Source |
| Release | 2025-12-15Source |
| Parameters | 30 BSource |
| Active parameters | 3.5 BSource |
| License | NVIDIA Nemotron Open Model LicenseSource |
| Architecture | Mamba2-Transformer hybrid MoESource |
| Languages | English, German, Spanish, French, Italian, JapaneseSource |
| 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
- 89.1
- Score
- 89.1
- 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
- 99.2
- Score
- 99.2
- 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
- 73
- Score
- 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. no tools
GPQA (Show measurement, test conditions, and source)
- Source value
- 75
- Score
- 75
- 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
LiveCodeBench v6 (Show measurement, test conditions, and source)
- Source value
- 68.3
- Score
- 68.3
- 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. 2025-08-2025-05 snapshot
MMLU-Pro (Show measurement, test conditions, and source)
- Source value
- 78.3
- Score
- 78.3
- 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-100@1M (Show measurement, test conditions, and source)
- Source value
- 86.3
- Score
- 86.3
- 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
Artificial Analysis Intelligence Index (Show measurement, test conditions, and source)
- Source value
- 6.8478519055181
- Score
- 6.8478519055181
- 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. Non-reasoning
AA-Omniscience Index (Show measurement, test conditions, and source)
- Source value
- -69.3
- Score
- -69.3
- 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. Non-reasoning
Output speed (Show measurement, test conditions, and source)
- Source value
- 222.599680250965
- Score
- 222.599680250965
- 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. Non-reasoning
- 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
- 0
- Score
- 0
- 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. Non-reasoning
- Context
- Artificial Analysis performance timing; output speed is decoded after the first streaming chunk. Provider basis: serving-provider median.
Latency input time (Show measurement, test conditions, and source)
- Source value
- 1.00139908700001
- Score
- 1.00139908700001
- 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. Non-reasoning
- Context
- Artificial Analysis performance timing; output speed is decoded after the first streaming chunk. Provider basis: serving-provider median.
Page 1 of 2
Our tests
SVG short test
One SVG prompt and fixed output settings. Captures are shown only with 0 reported reasoning tokens or documented lack of reasoning controls. Identical test conditions do not make the models' internal compute identical.
Earlier tests
Pelican on a bicycle
A community classic for free-form SVG drawing.
Nemotron 3 Nano
Output withheld for safety
Reasoning (requested): high
Current attempt
- Run date
- Oct 2, 2026
- API provider
- openrouter
- Requested API model
- nvidia/nemotron-3-nano-30b-a3b
- API model reported in response
- nvidia/nemotron-3-nano-30b-a3b
- Reasoning setting
- high
- Pinned endpoint
- Crusoe | nvidia/nemotron-3-nano-30b-a3b
- Temperature
- Provider default
- Token limit including reasoning
- 8,192 tokens
- Test protocol
- community-visual-tests-v1
- Technical checks
- Failed
- Response time
- 7.1 s
Not yet visually reviewed. A successful recording does not confirm correct physics or full compliance with the prompt.
Earlier attempts (1)
- Oct 1, 2026Output withheld for safetyReasoning setting: high
Prompt and test conditions
Generate an SVG of a pelican riding a bicycle
Token limit including reasoning: 8,192 tokens
Task origin (Simon Willison)Popularity
44.3 / 100
Predecessor and successor
Sources and data date
| Type | Evidence and data date |
|---|---|
| Speed | Median across serving providers, Non-reasoning. 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 | NVIDIA Nemotron Nano 2 9B model card · 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 |
| Additional source | NVIDIA Nemotron 3 Nano launch blog (Hugging Face) (2026-09-19) |