Nemotron Nano 9B is NVIDIA's open-weight 9B general-purpose reasoning model, built as a unified foundation for both reasoning and non-reasoning tasks. It uses a hybrid Mamba-2 Transformer architecture, made up primarily of Mamba-2 and MLP layers with just a few attention layers, for efficient inference while retaining strong reasoning performance. Trained on diverse multilingual and multi-code data covering six natural languages and 43 programming languages, the model supports 128K token context and reasoning-enabled inference through chain-of-thought generation. It handles math, code, science, and general-knowledge domains with a unified approach to instruction-following and problem-solving, and offers runtime control over its thinking budget. Built for developers creating cost-effective AI agents, chatbots, and tool-calling systems where 9B parameters balance capability and resource use, it suits responsiveness-sensitive applications. Reasoning can be toggled via the system prompt, with the model generating reasoning traces before its conclusions.
Nemotron Nano 9B runs on 1 route through the Opper gateway. Compare residency, ZDR, and training posture at a glance — full data-handling detail per route below.
| Provider | Region | Zero data retention | Training | Input | Output |
|---|---|---|---|---|---|
| AWS Bedrock | US | Zero data retention | No | $0.04 | $0.16 |
Training posture across routes: No training on prompts by default.
Each route hosting Nemotron Nano 9B has its own privacy posture, residency, and GDPR terms. Postures are maintained by Opper with a last-verification timestamp.
Zero data retention is on by default on Pay-as-you-go — no action required. No training on customer data. US; SCCs; DPA available.
Call Nemotron Nano 9B through the Opper gateway with one API key. Let your coding agent set it up, or call it directly — Opper is drop-in compatible with the OpenAI, Anthropic, and Google AI SDKs.
Copy this and paste it into your coding agent — Claude Code, Cursor, Codex, and more — and it'll wire up Opper for you.
import OpenAI from "openai";const client = new OpenAI({apiKey: process.env.OPPER_API_KEY,baseURL: "https://api.opper.ai/v3/compat",});const completion = await client.chat.completions.create({model: "aws/nemotron-nano-9b-v2",messages: [{ role: "user", content: "Hello" }],});console.log(completion.choices[0].message.content);
Side-by-side on privacy, EU hosting, pricing, and benchmarks.
