Requesty
Nebius AI logo
Nebius AI

nvidia/nemotron-3-ultra-550b-a55b

Nemotron 3 Ultra is a 550B hybrid MoE model from NVIDIA, optimized for the most demanding multi-agent AI and complex reasoning tasks.

ReasoningTool callingJSON schema
Try nvidia/nemotron-3-ultra-550b-a55bCompareDocs

Specifications

Context window128K tokens
Max outputN/A
API typechat
AddedJun 16, 2026
Model IDnebius/nvidia/nemotron-3-ultra-550b-a55b
Data retentionNo
Used for trainingNo
Provider location🇪🇺 EU
Privacy policyNebius Privacy Policy→

Benchmarks

Benchmarks haven't been published yet for this exact variant.

Some variants (region-specific deployments, highspeed tiers) share benchmarks with their base model. Check the base model page or the Nebius AI models overview.

Pricing

Input / 1M
$1.00
Output / 1M
$3.00
Cache write
N/A
Cache read
N/A
Estimated cost
100K input + 10K output$0.13
1M input + 100K output$1.30
10M input + 1M output$13.00

Requesty charges exactly what the upstream provider charges, no markup, no per-request fees. Prompt caching and smart routing can reduce effective cost by 30-80%.

Quickstart

Drop-in compatible with the OpenAI SDK. Change the base URL, swap in your Requesty API key, and set the model to nebius/nvidia/nemotron-3-ultra-550b-a55b.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_REQUESTY_API_KEY",
    base_url="https://router.requesty.ai/v1",
)

response = client.chat.completions.create(
    model="nebius/nvidia/nemotron-3-ultra-550b-a55b",
    messages=[
        {"role": "user", "content": "Explain quantum computing in one paragraph."},
    ],
)

print(response.choices[0].message.content)

Other Nebius AI models

glm-5.2
1M ctx$1.40/1M
Coding 51
gemma-3-27b-it
128K ctx$0.10/1M
qwen/qwen3-32b
128K ctx$0.10/1M
qwen/qwen3-next-80b-a3b-thinking
128K ctx$0.15/1M
nousresearch/hermes-4-70b
128K ctx$0.13/1M
qwen/qwen3-30b-a3b-instruct-2507
128K ctx$0.10/1M

Frequently asked questions

How much does nvidia/nemotron-3-ultra-550b-a55b cost?
nvidia/nemotron-3-ultra-550b-a55b is priced at $1.00 per million input tokens and $3.00 per million output tokens when accessed via Requesty. Requesty charges exactly what the upstream provider charges, we don't add markup.
What is the context window of nvidia/nemotron-3-ultra-550b-a55b?
nvidia/nemotron-3-ultra-550b-a55b has a context window of 128K tokens. That's roughly 171 words of input you can fit in a single prompt.
What can nvidia/nemotron-3-ultra-550b-a55b do?
nvidia/nemotron-3-ultra-550b-a55b supports tool calling, extended reasoning, structured outputs (JSON schema). You can call it through any OpenAI-compatible client by pointing base_url to Requesty.
How do I use nvidia/nemotron-3-ultra-550b-a55b with the OpenAI SDK?
Install the OpenAI SDK, set base_url to "https://router.requesty.ai/v1", set your API key to your Requesty key, and set the model to "nebius/nvidia/nemotron-3-ultra-550b-a55b". The Quickstart above shows Python, JavaScript and cURL snippets.
Can I run nvidia/nemotron-3-ultra-550b-a55b through Requesty?
Yes. nvidia/nemotron-3-ultra-550b-a55b runs through Requesty's OpenAI-compatible API, served from Nebius AI. You do not host the model yourself: point base_url at Requesty, set the model to "nebius/nvidia/nemotron-3-ultra-550b-a55b", and requests are routed to the upstream provider with automatic failover. The same key gives you 400+ other models too.

Access nvidia/nemotron-3-ultra-550b-a55b through Requesty

One API key, 400+ models, OpenAI-compatible. No markup on provider prices, automatic failover, and smart caching built-in.

Get started freeAll Nebius AI models