NVIDIA logo

Nemotron 3 Ultra 550B A55B

Created by

NVIDIA

Release date:Aug 20, 2026nvidia/nvidia-nemotron-3-ultra-550b-a55b
Release date:Aug 20, 2026nvidia/nvidia-nemotron-3-ultra-550b-a55b

NVIDIA Nemotron 3 Ultra is a 550B hybrid MoE model for multi-agent and complex reasoning workloads.

NVIDIA Nemotron 3 Ultra is a 550B hybrid MoE model for multi-agent and complex reasoning workloads.

Get your API key

Read the docs

Compare with another model

Context window

1,000,000 tokens

Modalities

Input:

Text

Output:

Text

Price from

$0.25 input · $2.50 output
/1M tokens via Perplexity (Global)

Benchmarks

Coding 49.3 · Intelligence 22.9

Artificial Analysis indexes. Same model on every route.

Capabilities

Available across listed offers. Support varies by route.

Reasoning

Tool calling

JSON mode

Streaming

Providers & pricing

Compare every listed route. Open route details for model IDs, endpoints, service tiers and all pricing conditions.

Provider
Region
Input
Output
Cached input
Zero data retention
Route ID
Perplexity
Lowest price
Global
$0.25/1M tokens
$2.50/1M tokens
$0.25/1M tokens
?
perplexity/perplexity/nemotron-3-ultra-550b-a55b
NVIDIA
Global
Not listed
Not listed
Not listed
?
nvidia/nvidia/nemotron-3-ultra-550b-a55b
Nebius
Global
$1.00/1M tokens
$3.00/1M tokens
Not listed
?
nebius/nvidia/Nemotron-3-Ultra-550b-a55b
Baseten
Global
$0.60/1M tokens
$2.40/1M tokens
$0.12/1M tokens
?
baseten/nemotron-ultra
Provider
Region
Input
Output
Cached input
Zero data retention
Route ID
Perplexity
Lowest price
Global
$0.25/1M tokens
$2.50/1M tokens
$0.25/1M tokens
?
perplexity/perplexity/nemotron-3-ultra-550b-a55b
NVIDIA
Global
Not listed
Not listed
Not listed
?
nvidia/nvidia/nemotron-3-ultra-550b-a55b
Nebius
Global
$1.00/1M tokens
$3.00/1M tokens
Not listed
?
nebius/nvidia/Nemotron-3-Ultra-550b-a55b
Baseten
Global
$0.60/1M tokens
$2.40/1M tokens
$0.12/1M tokens
?
baseten/nemotron-ultra
Provider
Region
Input
Output
Cached input
Zero data retention
Route ID
Perplexity
Lowest price
Global
$0.25/1M tokens
$2.50/1M tokens
$0.25/1M tokens
?
perplexity/perplexity/nemotron-3-ultra-550b-a55b
NVIDIA
Global
Not listed
Not listed
Not listed
?
nvidia/nvidia/nemotron-3-ultra-550b-a55b
Nebius
Global
$1.00/1M tokens
$3.00/1M tokens
Not listed
?
nebius/nvidia/Nemotron-3-Ultra-550b-a55b
Baseten
Global
$0.60/1M tokens
$2.40/1M tokens
$0.12/1M tokens
?
baseten/nemotron-ultra

Put your chosen AI models to work with Orq.

Compare providers, capabilities, and pricing across the live Orq.ai catalog.

Browse all models

Start routing

Book a briefing