Skip to main content
NVIDIAJun 4, 2026

NVIDIA Nemotron 3 Ultra

nvidia-nemotron-3-ultra-550b-a55bPrivateOpen weights

NVIDIA Nemotron 3 Ultra is built for frontier reasoning, orchestration, coding agents, deep research, and complex enterprise workflows. It delivers up to 5x faster inference and up to 30% lower cost for agentic workloads.

$0.625/$3.13
per 1M in / out
256K
context window
32K
max output
How it comparesamong 143 text models
GPQA Diamondhigher than 33%79.8%
Blended pricecheaper than 58%$1.25
Context windowlarger than 22%256K
Max outputlarger than 21%32K

Further right is better. Hover a dot for its model.

Time to first token— 
Output speed— 
Uptime— 

Last 24 hours of Venice traffic.How it’s measured

Capabilities

What the model takes in, returns and supports through the API.

InText
OutText

Not supportedimage input, video input, audio input, X search and log probabilities

Benchmarks

GPQA Diamond
79.8%91st of 139 · 24 score higher for less, including DeepSeek v4.1 Flash (1017), MiniMax M2.5 and DeepSeek V4.1 Flash 1017 Uncensored

Specs

Model ID
nvidia-nemotron-3-ultra-550b-a55b
Endpoint
POST /api/v1/chat/completions
Context window
256,000 tokens
Max output
32,768 tokens
Reasoning
none · low · medium · high · default low
Precision
FP8
Privacy
Private
Added
Jun 4, 2026

Compare

ModelInputOutputGPQA DiamondContextMax outputFirst tokenSpeed
NVIDIA Nemotron 3 Ultra$0.625$3.1379.8%256K32K——
Grok Build 0.1$1.00$2.0089.6%256K64K——
Seed 2.1 Turbo$0.625$3.1370.7%256K64K——
Qwen 3.8 27B E2EE$0.47$3.5343.6% (off)256K8K——
Kimi K2.7 Code$0.75$3.5084.3%256K64K——

Pricing

Billed per token in USD. Reasoning tokens bill as output.

Input$0.625per 1M tokens
Cached input$0.187570% off input
Output$3.13per 1M tokens

Estimate a workload

$0.0028per request
$28.13for 10,000 requests

Blended price against similar models

NVIDIA Nemotron 3 Ultra$1.25

Per 1M tokens at three input tokens for each output token.

Quickstart

OpenAI-compatible: point any OpenAI SDK at the Venice base URL with a Venice API key.

curl https://api.venice.ai/api/v1/chat/completions \
-H "Authorization: Bearer $VENICE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "nvidia-nemotron-3-ultra-550b-a55b",
"messages": [{ "role": "user", "content": "Explain TEE attestation in two sentences." }],
"reasoning_effort": "low"
}'
Base URL https://api.venice.ai/api/v1

Parameters

ParameterValuesNotes
modelnvidia-nemotron-3-ultra-550b-a55bRequired
messagestextContent parts the model reads
max_completion_tokensUp to 32,768Counts reasoning tokens
reasoning_effortnone | low | medium | highDefault low
toolsFunction definitionsWith tool_choice
response_format{ type: "json_schema" }Output that matches a schema
streamtrue | falseServer-sent events
venice_parameters.enable_web_searchauto | on | offWeb results and citations