Nemotron 3 Ultra
nvidia/nemotron-3-ultra-550b-a55b
- context
- 262K
- modalities
- text → text
- capabilities
- Tool use, Reasoning, Structured output
Overview
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...
Providers
- BaseTenView BaseTen
- DeepInfraView DeepInfra
- VeniceView Venice
Access method depends on provider and account setup.
Pricing
| provider | input / 1M | output / 1M |
|---|---|---|
| BaseTen | $0.60 | $2.40 |
| DeepInfra | $0.50 | $2.20 |
| Venice | $0.625 | $3.125 |
BaseTen
- input / 1M
- $0.60
- output / 1M
- $2.40
DeepInfra
- input / 1M
- $0.50
- output / 1M
- $2.20
Venice
- input / 1M
- $0.625
- output / 1M
- $3.125
Use this model
through Nautili
Route requests to Nemotron 3 Ultra through your Nautili gateway using a supported SDK or the Nautili SDK.
Connect your stack to Nautili
Route across supported providers through one control layer.