Model · NVIDIA

Nemotron 3 Ultra

nvidia/nemotron-3-ultra

Try in Agents

NVIDIA's 550B open reasoning model (55B active), a hybrid Mamba-Transformer MoE built for long-running agent workflows

reasoningcaching

Specs

Context

Max output

16K

Provider

NVIDIA

Input

$0.600/ 1M tokens

Output

$2.40/ 1M tokens

Cache read

$0.120/ 1M tokens

Try it

Chat with Nemotron 3 UltraTerramind API

⌘+Enter to run

Code

import { streamText } from "ai";
import { createOpenAICompatible } from "@ai-sdk/openai-compatible";

const terramind = createOpenAICompatible({
  name: "terramind",
  baseURL: "https://terramind.com/api/v1",
  apiKey: process.env.TERRAMIND_API_KEY!,
});

const result = streamText({
  model: terramind("nemotron-3-ultra"),
  prompt: "Why is the sky blue?",
});

for await (const chunk of result.textStream) {
  process.stdout.write(chunk);
}

Want to run this in your app? Grab an API key · Full chat reference

More by NVIDIA

Model	Context	In	Out	Capabilities
nvidia/nemotron-3-super	256K	$0.150	$0.650	reasoning
nvidia/nemotron-3.5-lightning	262K	$0.050	$0.200	reasoningcaching