# Nemotron 3 Ultra — price & specs

Source: https://modelbank.ai/models/nemotron-3-ultra-550b-a55b · updated 2026-11-12 · ModelBank

| Field | Value |
|---|---|
| Vendor | NVIDIA |
| Input price | $0.6 / 1M tokens |
| Output price | $2.4 / 1M tokens |
| Cached input | $0.12 / 1M tokens |
| Context window | 262,144 tokens |
| Max output | 182,520 tokens |
| Input modalities | text |
| Reasoning / tool use | yes / yes |
| Intelligence / coding index | 11.8 / 49.3 |
| Released | 2026-06-04 |
| Retirement date | none announced |

## About

NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE). Built on a hybrid Transformer-Mamba mixture-of-experts architecture, it...

## Price history

- 2026-09-24: $0.6 in / $2.4 out
- 2026-10-02: $0.5 in / $2.2 out
- 2026-10-02: $0.6 in / $2.4 out
- 2026-10-02: $0.5 in / $2.2 out
- 2026-10-11: $0.6 in / $2.4 out

## Cheaper alternatives with similar intelligence

- [GLM 5.3 Flash](https://modelbank.ai/models/glm-5.3-flash): $0.0399 / $0.0784, intelligence 27.8
- [DeepSeek V4.1 Flash 1017](https://modelbank.ai/models/deepseek-v4.1-flash-1017): $0.098 / $0.29, intelligence 28.6
- [MiMo-V2.6-Flash](https://modelbank.ai/models/mimo-v2.6-flash): $0.14 / $0.28, intelligence 26.7
- [Claude Haiku 5.5](https://modelbank.ai/models/claude-haiku-5.5): $0.1 / $0.5, intelligence 36.5
- [GPT-6 Luna](https://modelbank.ai/models/gpt-6-luna): $0.1 / $0.5, intelligence 27.9

JSON: https://modelbank.ai/api/v1/models/nemotron-3-ultra-550b-a55b