TokenRate

Nemotron 3 Ultra Pricing

Flagship

NVIDIA · 1M tokens context

Nemotron 3 Ultra from NVIDIA costs $0.500 per 1 million input tokens and $2.20 per 1 million output tokens as of July 2026 (live OpenRouter data). The model supports a 1,000,000-token context window (approximately 750,000 words) with a 16K-token maximum output. A typical 1,000-token request costs $0.0005 in input charges; a 10,000-token request costs $0.0050.

Nemotron 3 Ultra pricing and capability summary
Input price$0.500 / 1M tokens
Output price$2.20 / 1M tokens
Output / input ratio4.4×
Context window1,000,000 tokens (~750,000 words)
Maximum output16,384 tokens
Cost per 1K tokens (input)$0.0005
TierFlagship
Last verified

NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE).

Live pricing from OpenRouter

Input Price

$0.500

per 1 million tokens

Output Price

$2.20

per 1 million tokens

Context Window

1M tokens

max 16K output

Cost Examples

Request TypeTokensInput CostOutput Cost
1,000 word article1,333$0.000666$0.00088
10-page document (2,500 words)3,333$0.00167$0.0022
1,000 lines of code5,000$0.0025$0.0033
100K token document100,000$0.05$0.066

Output cost estimated at 30% of input token count. Use the calculator for exact figures.

Strengths

  • Affordable at $0.50/1M input tokens
  • Massive 1M-token context window
  • Frontier-class quality on complex tasks

Limitations

  • Quality and availability can vary by hosting provider
  • Quality and availability can vary by hosting provider

Best Use Cases

Complex research and analysis
Advanced coding and architecture
Long-form content generation
High-stakes production workloads

Calculate Nemotron 3 Ultra Costs

Use the TokenRate calculator to convert any budget, token count, or text into exact Nemotron 3 Ultra costs — and compare across all models.

Open Calculator →

Nemotron 3 Ultra — FAQ

Related Models

Related Guides