All Models

Nemotron 3 Ultra

nemotronReasoningTool CallingOpen WeightsStructured Output

Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy

Providers8
ReleasedJun 4, 2026
Input Modalitiestext
Output Modalitiestext
Tarsk Usecoding

Benchmarks

23.4
Intelligence Index
49.3
Coding Index

Available Providers (8)

ProviderModel IDInput CostOutput CostContextMax OutputDocs
Bothubnemotron-3-ultra-550b-a55b:free$0/MTok$0/MTok1M128K
Kilo Gatewaynvidia/nemotron-3-ultra-550b-a55b:free$0/MTok$0/MTok1M65.5K
OpenRouternvidia/nemotron-3-ultra-550b-a55b:free$0/MTok$0/MTok1M65.5K
Ollama Cloudnemotron-3-ultra$0.10/MTok$3/MTok262.1K128K
CoreWeavenvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B$0.50/MTok$2.15/MTok262.1K262.1K
Vercel AI Gatewaynvidia/nemotron-3-ultra-550b-a55b$0.60/MTok$2.40/MTok1M65K
Venice AInvidia-nemotron-3-ultra-550b-a55b$0.63/MTok$3.13/MTok256K32.8K
DigitalOceannemotron-3-ultra-550b$0.90/MTok$1.70/MTok131.1K131.1K

Capabilities

Reasoning
Tool Calling
Attachments
Open Weights
Structured Output