Nemotron 3 Ultra
By nvidia
nvidia/nemotron-3-ultra-550b-a55bA 550B parameter (55B active) open reasoning model from NVIDIA, built for long-running agent workflows. It uses a hybrid Mamba-Transformer MoE architecture and supports a 1M token context window.
ReleasedJun 4, 2026
Pricing
Pricing (Input)$0.60/M
Pricing (Output)$2.40/M
CachingYes
Limits
Context Length1M
Max Output Tokens65K
Input Modalities
TextSupported
ImageNot supported
PDFNot supported
AudioNot supported
Output Modalities
TextSupported
ImageNot supported
AudioNot supported
Features
ReasoningSupported
ToolsSupported
TemperatureSupported