Does Nemotron 3 Nano Omni support zero data retention?

Not self-serve on Pay-as-you-go. Zero data retention for Nemotron 3 Nano Omni is available on request via Opper Enterprise — contact hello@opper.click.

Where is Nemotron 3 Nano Omni hosted?

Nemotron 3 Nano Omni runs on EU hosting routes through the Opper gateway, with contractually enforced residency on supported providers.

Is Nemotron 3 Nano Omni trained on prompts and outputs?

No. Every hosting route for Nemotron 3 Nano Omni is configured so customer prompts and outputs are not used to train models by default.

Is a GDPR DPA available for Nemotron 3 Nano Omni?

Yes. A GDPR Data Processing Agreement is available on at least one route serving Nemotron 3 Nano Omni. Contact hello@opper.click to receive it.

Nemotron 3 Nano Omni

by NVIDIA

Nemotron 3 Nano Omni unifies vision, audio, language, and video understanding in a single 30B-parameter open-weight model with about 3B active parameters per token. The hybrid Mamba-Transformer MoE architecture enables efficient multimodal processing while offering 65K token context and structured outputs for agentic workflows. The model natively processes images, video, audio, and text, with capabilities including optical character recognition, GUI understanding, speech transcription, and video captioning, making it well suited to document intelligence and visual reasoning. Designed for enterprises building agents that work with visual and temporal data, it targets customer service automation, document processing, content analysis, and screen-based task automation. The model is available in BF16, FP8, and NVFP4 quantizations for flexible hardware deployment.

Open in playground Try in chat Compare

Key info

Input

Output

Features

Context window

66K

Max output

—

Input price

$0.06 /1M

Output price

$0.24 /1M

EU residency available
Zero data retention via Enterprise
No training by default
GDPR DPA available

Available routes

Nemotron 3 Nano Omni 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
Nebius	EU	Enterprise	No	$0.06	$0.24

Training posture across routes: No training on prompts by default.

Data handling per route

Each route hosting Nemotron 3 Nano Omni has its own privacy posture, residency, and GDPR terms. Postures are maintained by Opper with a last-verification timestamp.

Nebius — Finland🇫🇮

Zero data retention is available via Opper Enterprise contract. No training on customer data. EU; DPA available.

Zero data retention: Available via Opper Enterprise contract.
Training: No training on customer data.
Logging: Abuse monitoring
Third-party access: None disclosed
GDPR DPA: DPA available
Transfer mechanism: Not applicable — data stays in EU

Benchmarks

Independent benchmark scores — composite indices for reasoning, coding, and math, plus individual eval scores where available.

Global rank#253 of 534 LLMs

TierEfficient

Output speed296 tok/s

First token0.59s

Intelligence Index14.9

Reasoning & knowledge

GPQA Diamond

47%

Humanity's Last Exam

Long-context reasoning

36%

Coding

SciCode

28%

Agentic & tool use

Terminal-Bench Hard

τ²-Bench Telecom

45%

Math & instruction following

IFBench

63%

See full leaderboard →Benchmarks via Artificial Analysis · View on AA

Get started

Call Nemotron 3 Nano Omni 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.

Set it up with your agent

Copy this and paste it into your coding agent — Claude Code, Cursor, Codex, and more — and it'll wire up Opper for you.

Or call it directly

1import OpenAI from "openai";
2
3const client = new OpenAI({
4  apiKey: process.env.OPPER_API_KEY,
5  baseURL: "https://api.opper.click/v3/compat",
6});
7
8const completion = await client.chat.completions.create({
9  model: "nebius/nvidia/Nemotron-3-Nano-Omni",
10  messages: [{ role: "user", content: "Hello" }],
11});
12console.log(completion.choices[0].message.content);