nvidia

Nemotron 3 Ultra

A 550B parameter (55B active) open reasoning model from NVIDIA, built for long-running agent workflows. It uses a hybrid Mamba-Transformer MoE architecture and supports a 1M token context window.

text inputtext outputlanguage
const response = await fetch('https://api.minirouter.sh/v1/chat/completions', {  method: 'POST',  headers: {    Authorization: `Bearer ${process.env.MINIROUTER_KEY}`,    'Content-Type': 'application/json',  },  body: JSON.stringify({    model: 'nvidia/nemotron-3-ultra-550b-a55b',    messages: [{ role: 'user', content: 'Why is the sky blue?' }],  }),}) const data = await response.json()console.log(data.choices[0].message.content)
Read docs →

Overview

Model type
language
Context window
1,000,000
Maximum output
65,000
Input / 1M tokens
$0.63
Output / 1M tokens
$2.52
Released
2026-06-04

Prices include the MiniRouter fee. Compatibility snapshot: 2026-08-06. Input: text. Output: text.

Endpoints

/v1/chat/completions/v1/messages/v1/responses

Platform-funded routing only. Caller BYOK credentials and OIDC-based upstream authentication are intentionally excluded; authenticate with a MiniRouter bearer key.

Captured parameters: max_tokens, temperature, stop, tools, tool_choice, reasoning, include_reasoning.

API

Use the setup panel above to switch between Chat Completions, Messages, and client configurations. Every snippet keeps the model ID unchanged.

Base URLhttps://api.minirouter.sh/v1
Model IDnvidia/nemotron-3-ultra-550b-a55b

Providers

MiniRouter selects from the configured routes below. Priority is the catalog order, not a latency or availability score.

ProviderPriorityInput / 1MOutput / 1MParametersAccess
vercel
1$0.63$2.52max_tokens, temperature, stop, tools, tool_choice, reasoning, include_reasoningMiniRouter balance