Model Details
Released
Jun 4, 2026
Context Length
1,048,576
Where is Nemotron-3 Ultra-550b-a55b a perfect fit?
NVIDIA Nemotron 3 Ultra 550B A55B is NVIDIA's largest Nemotron 3 model, built for advanced reasoning, agentic workflows, software engineering, and long-context analysis. It uses a hybrid Mamba-Transformer Mixture-of-Experts architecture with 550B total parameters and 55B active parameters per token.
- Complex mathematical, scientific, and technical reasoning
- Large-scale software engineering and codebase analysis
- Autonomous multi-step agents
- Tool calling and agentic workflow execution
- Long-context document and knowledge-base analysis
- Technical research and enterprise knowledge work
- Code generation, debugging, and refactoring
- Self-hosted deployments where open weights and infrastructure control are important
- Complex mathematical, scientific, and technical reasoning
- Large-scale software engineering and codebase analysis
- Autonomous multi-step agents
- Tool calling and agentic workflow execution
- Long-context document and knowledge-base analysis
- Technical research and enterprise knowledge work
- Code generation, debugging, and refactoring
- Self-hosted deployments where open weights and infrastructure control are important
Quick Model Estimate
Your Nemotron-3 Ultra-550b-a55b Cost Estimate
💰 Total Cost
—
for 1000 input + 1000 output tokens
📥 Input (1000 × $0.250000)
—
📤 Output (1000 × $2.500000)
—
Cost Breakdown
📥 Input
📤 Output
Prices are watched for changes and checked against the provider's own page.
Pricing
|
Provider
↕
|
Modality
↕
|
Service Tier
↕
|
Input Price
(per 1M tokens) ↕ |
Output Price
(per 1M tokens) ↕ |
Cached Input
(per 1M tokens) ↕ |
Context Size
↕
|
View
|
|---|---|---|---|---|---|---|---|
 Perplexity | Text | Standard | $0.2500 | $2.5000 | $0.2500 | 1,048,576 tokens | → |
Other Models in the Nemotron-3 Family
FAQs about Nemotron-3 Ultra-550b-a55b
How much does Nemotron-3 Ultra-550b-a55b cost per 1M tokens?
Nemotron-3 Ultra-550b-a55b costs $0.25 per million input tokens, and $2.50 per million output tokens.
What does a typical workload cost with Nemotron-3 Ultra-550b-a55b?
1,000 requests of 2,000 input and 500 output tokens each — 2,000,000 input and 500,000 output tokens in total — costs $1.75 with Nemotron-3 Ultra-550b-a55b at its lowest rates. Use the calculator on this page for your own volumes.
Where can I use Nemotron-3 Ultra-550b-a55b?
Nemotron-3 Ultra-550b-a55b is available through Perplexity, from $0.25 per million input tokens.
What is Nemotron-3 Ultra-550b-a55b's context window?
Nemotron-3 Ultra-550b-a55b has a context window of 1,048,576 tokens.