Nemotron 3 Ultra 550B A55B
No independent test. This API route aliases nemotron-3-ultra.
Language model
Nemotron 3 Ultra 550B A55B
NVIDIA
- Released
- June 4, 2026
- Data date
- September 26, 2026
- Input
- Text
- Output
- Text
Nemotron 3 Ultra 550B A55B is NVIDIA's BF16 checkpoint in the Ultra tier. Its hybrid LatentMoE architecture combines Mamba2, MoE, and attention with multi-token prediction. It has 550 billion total parameters and activates 55 billion.
NVIDIA advertises up to 1M context tokens. Its documented vLLM and SGLang examples start at 262,144 tokens and require an additional long-context switch for the larger window.
Specifications and access
| Specification | Value and source |
|---|---|
| Weights | Open weightsSource |
| Access | APISource |
| Parameters | 550 billionSource |
| Active parameters | 55 billionSource |
| License | OpenMDW-1.1Source |
| Inputs | TextSource |
| Outputs | TextSource |
| Reasoning | YesSource |
| Tool calling | YesSource |
| Advertised context | Up to 1M tokensSource |
| Packaged default configuration | 262,144 tokensSource |
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 Ultra 550B A55B
Reasoning (requested): high
Current attempt
- Run date
- Oct 1, 2026
- API provider
- openrouter
- Requested API model
- nvidia/nemotron-3-ultra-550b-a55b
- API model reported in response
- nvidia/nemotron-3-ultra-550b-a55b
- Reasoning setting
- high
- Pinned endpoint
- DeepInfra | nvidia/nemotron-3-ultra-550b-a55b-20260604
- Temperature
- Provider default
- Token limit including reasoning
- 8,192 tokens
- Test protocol
- community-visual-tests-v1
- Technical checks
- Passed, not a quality rating
- Response time
- 39.4 s
Visually reviewed. This individual test does not establish overall model quality.
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
20.4 / 100
Sources and data date
| Type | Evidence and data date |
|---|---|
| Input | Documented modalities. Retrieved September 26, 2026. |
| Output | Documented modalities. Retrieved September 26, 2026. |
| Additional source | NVIDIA Nemotron 3 Ultra 550B A55B model card (retrieved October 2, 2026) · NVIDIA Nemotron 3 Ultra BF16 model card · Editorial description reviewed October 2, 2026 |
| Additional source | NVIDIA Nemotron 3 Ultra 550B A55B packaged config (retrieved October 2, 2026) |
| Additional source | Nemotron 3 Ultra 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 |