Back to models

NVIDIA: Nemotron 3 Ultra

NVIDIA · Text

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

Model ID nvidia/nemotron-3-ultra-550b-a55bUpdated Jun 4, 2026

Pricing & billing

Billing unit: 1M tokens. Vendor reference price; conversion is approximate.

ItemUSD referenceList price (USD)
Input$0.50$0.50
Output$2.20$2.20
Cache read—No public pricing—No public pricing
Cache write—No public pricing—No public pricing

Specifications

Form
Text model
Input modalities
Text
Output modalities
Text
Context length
262K
Max output
16K
Weights
Open weights
Family
nemotron
Knowledge cutoff
2025-09
Released
Jun 4, 2026

Uses and capabilities

Capabilities Text model、Function calling、Structured output、Reasoning、Open weights

Function calling

Structured output

Reasoning

Prompt caching

Open weights

Sources

Use these links to verify specs and pricing — they may point to the vendor's site or to a third-party catalog.