# Nemotron Ultra

> Nemotron Ultra is an open-weight language model from NVIDIA, released on 4 Jun 2026. It accepts text and generates text, with a 203K-token context window and up to 203K output tokens per response. As of 4 Oct 2026, the lowest listed API price is $0.50 per million input tokens and $2.20 per million output tokens (OpenRouter), across 7 providers we track.

Source page: https://www.aimodel.directory/models/nemotron-3-ultra-550b-a55b
Last verified: 4 Oct 2026

## Specifications

- **Developer:** NVIDIA
- **Release date:** 4 Jun 2026 (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/baseten/models/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B.toml)
- **Status:** Generally available (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/baseten/models/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B.toml)
- **Weights:** Open weights (source: Hugging Face, https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16)
- **License:** openmdw-1.1 (source: Hugging Face, https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16)
- **Parameters:** 561B (source: Hugging Face, https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16)
- **Context window:** 203K (202,800 tokens) (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/baseten/models/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B.toml)
- **Max output:** 203K tokens (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/baseten/models/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B.toml)
- **Knowledge cutoff:** Feb 2026 (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/baseten/models/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B.toml)
- **Input:** Text (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/baseten/models/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B.toml)
- **Output:** Text (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/baseten/models/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B.toml)
- **Reasoning mode:** Yes (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/baseten/models/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B.toml)
- **Tool calling:** Yes (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/baseten/models/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B.toml)
- **Structured output:** Yes (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/baseten/models/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B.toml)
- **Hugging Face:** nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 (source: Hugging Face, https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16)

## API pricing (USD per 1M tokens)

| Provider | Input | Output | Cached input | Context |
|---|---|---|---|---|
| NVIDIA NIM | $0.50 | $2.50 | $0.15 | 1M |
| OpenRouter | $0.50 | $2.20 | $0.10 | 262K |
| Fireworks AI | $0.60 | $2.40 | $0.12 | 262K |
| Baseten | $0.60 | $2.40 | $0.12 | 203K |
| Together AI | $0.60 | $3.60 | $0.20 | 512K |
| Vercel AI Gateway | $0.60 | $2.40 | $0.12 | 1M |
| Nebius AI Studio | $1 | $3 | $1 | 1M |

## FAQ

### What is the context window of Nemotron Ultra?

Nemotron Ultra has a context window of 202,800 tokens (203K) and can generate up to 202,800 tokens in a single response, according to models.dev.

### How much does the Nemotron Ultra API cost?

Nemotron Ultra is listed from $0.50 per million input tokens and $2.20 per million output tokens (OpenRouter). Prices last verified 4 Oct 2026.

### Is Nemotron Ultra open source?

Nemotron Ultra is an open-weight model: its weights are published on Hugging Face as nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 under the openmdw-1.1 license. Check the license terms before commercial use.

### When was Nemotron Ultra released?

NVIDIA released Nemotron Ultra on 4 Jun 2026, according to models.dev.

### Which providers offer Nemotron Ultra?

We track Nemotron Ultra on 7 providers: NVIDIA NIM, OpenRouter, Fireworks AI, Baseten, Together AI, Vercel AI Gateway, Nebius AI Studio.

---
Data licensed CC BY 4.0. Cite as: AI Model Directory, "Nemotron Ultra", https://www.aimodel.directory/models/nemotron-3-ultra-550b-a55b