NewChat Models

NVIDIA

Nemotron 3 Nano

Compact Nemotron model for short replies, extraction and bounded reasoning, using a paid 256K OpenRouter route.

CompactReasoningText
Open AppBrowse All Models

At a glance

Know the model before you prompt.

OpenRouter API specifications
Context window
262,144 tokens
Maximum output
Provider-dependent
Inputs → output
Text → Text
Knowledge cutoff
Not verified

API model ID: nvidia/nemotron-3-nano-30b-a3b

Capabilities & boundaries

What it supports. Where the limits are.

Tool support requires the appropriate API integration; a supported tool is not automatically active in every chat.

Supported API features and tools

  • A compact hybrid mixture-of-experts model: NVIDIA reports 30B total and 3.5B active parameters despite the A3B shorthand in the ID
  • Text-focused reasoning and non-reasoning tasks on the paid OpenRouter route
  • Conversational assistance and bounded code or document transformations
  • Tool calls and structured outputs on compatible endpoints such as the reviewed Nebius route

Before you choose

  • Compact size does not guarantee low latency or reliable handling of every difficult task. Use acceptance tests and escalate ambiguous or higher-risk results for review.
  • NVIDIA reports pre-training through June 25, 2025 and post-training through November 28, 2025. Those distinct dates are not one universal knowledge cutoff, and the model is not a live information feed.
  • The OpenRouter catalog reference is 262,144 tokens (256K). NVIDIA’s underlying model card describes up to 1M in suitable deployments; that is not the app’s confirmed route limit. Individual hosts can impose smaller windows. Leave room for the reply and reasoning.
  • This route accepts text and returns text, not image, audio or video input/output. Provide extracted text for document work; a prompt cannot add vision support.
  • Tool calls and structured outputs require a compatible host and integration. The model does not independently browse the web, execute code or access private files.
  • Answers, extracted fields and generated code can be wrong. Supply evidence and acceptance criteria, check cited passages and test code before use. A long context does not guarantee complete recall.

Extended thinking

NVIDIA describes both reasoning and non-reasoning use, controlled through a chat template in its own deployment examples. OpenRouter lists reasoning support for this paid route, but the reviewed catalog does not establish a shared default or named effort scale. Confirm the host and app controls before planning around them.

  • Use nvidia/nemotron-3-nano-30b-a3b without a :free suffix. A similarly named free endpoint is outside this app mapping and can have different limits.
  • Nebius lists 262,144 context and 235,929 maximum output tokens. Novita lists the same context and 32,768 output. A host ceiling does not guarantee a long answer or identical app allowance.
  • The reviewed Nebius endpoint lists structured_outputs, while Novita lists response_format without the same explicit structured_outputs capability. Validate schema support on the selected route.

Put it to work

Start with a more useful prompt.

Original examples from EZ Ai Assist. Adapt these to your task and the features available in your workspace.

Workflow 01

Classify short incoming messages

Keep an everyday task bounded and easy to audit.

Assign each message below to exactly one of the supplied categories. Return message ID, category, a short supporting phrase and needs_review as true or false. Use needs_review for ambiguous messages or missing context. Do not infer private account details and do not create new categories without permission.

Workflow 02

Rewrite a concise customer reply

Improve tone without changing facts or promises.

Rewrite this draft as a helpful reply under 90 words. Preserve every factual statement and the stated next step, remove jargon and do not add guarantees, discounts or deadlines. If the draft conflicts with the supplied policy, flag the conflict separately instead of silently choosing an interpretation.

Workflow 03

Extract a simple checklist

Separate stated tasks from assumptions.

Convert the notes below into a checklist using only explicit actions. For each action include its source sentence, owner if named and deadline if stated. Use not stated for missing values, keep suggestions separate from commitments and identify duplicates. Do not infer completion or assign work to someone who was not named.

Developer reference

OpenRouter API pricing

These are named-provider OpenRouter API reference prices, not EZ Ai Assist subscription prices.

View EZ Ai Assist plans
Nebius route reference on OpenRouter · USD per 1,000,000 tokens
Token typePrice
Input$0.06
Output$0.24
  • Rates were checked October 8, 2026. The named host is a pricing reference, not confirmation that the app pins that host. Routing, availability, cache eligibility and promotions can change the actual bill.
  • Nebius lists $0.06 input and $0.24 output per million tokens. No cache rate is assumed where the reviewed endpoint does not publish one.
  • Novita lists $0.05 input and $0.20 output with a different output ceiling. These are paid-provider examples; rates and limits for any :free endpoint are not used.
  • For a rough uncached estimate, multiply input tokens by the input rate and output tokens by the output rate, then divide by one million. Inspect the provider’s usage and billing records for reasoning and any additional charges.

Common questions

A few things worth knowing.

What is Nano best suited to testing first?

Start with bounded classification, concise replies and extraction tasks that have explicit output rules. Check failure cases before expanding the workload, and compare a larger model when the task needs more difficult reasoning.

Why does the ID say A3B while the card says 3.5B active?

The exact OpenRouter identifier uses the A3B shorthand. NVIDIA’s official Nano card reports 30B total and 3.5B active parameters. Keep the ID unchanged in integrations and use the card’s more specific figure when discussing architecture.

Is this the free Nemotron Nano endpoint?

No. The app owner confirmed nvidia/nemotron-3-nano-30b-a3b without :free. The pricing examples here are paid Nebius and Novita routes; free-route availability, quotas and rate limits do not describe this mapping.

Should I choose Nano or Lightning for short tasks?

Evaluate both on the same prompt set and acceptance criteria. Their rates, reasoning behavior and hosts differ; compact-model positioning alone does not establish the faster or cheaper end-to-end result.

Are these the app’s confirmed model IDs?

Yes. The app owner confirmed these exact paid OpenRouter IDs, with text-only 256K routes and no :free variants. That confirmation does not test a live request or guarantee that every hosting feature and reasoning control is exposed in your workspace.

Are the token prices EZ Ai Assist subscription prices?

No. These are named-host developer reference rates on OpenRouter. Hosting choices, discounts and supported features can change; your EZ Ai Assist plan and any limits displayed in the app are separate.

Can I use this model to inspect an image or search the web?

Not from the model alone. These routes are text-only. Supply extracted text or use a verified vision model for images. Web search and other tools require an enabled integration; a generated tool request is not proof an action happened.

Check the source

Official documentation

Specifications and API prices checked on . Example prompts and workflow advice are editorial guidance from EZ Ai Assist.

Same provider

More from NVIDIA

NVIDIA

Nemotron 3.5 Super VL

NVIDIA’s vision-language reasoner for text, images and visual documents. The app uses paid OpenRouter hosting; context limits vary by serving host.

NVIDIA

Nemotron 3 Ultra

NVIDIA’s flagship Nemotron reasoner for complex text, code and multi-step analysis, using a paid 256K OpenRouter route.

NVIDIA

Nemotron 3 Super

Balanced Nemotron workhorse for text, coding and tool-assisted workflows, using a paid 256K OpenRouter route.

NVIDIA

Nemotron 3.5 Lightning

Efficiency-focused Nemotron model for low-cost text workflows and rapid iteration, using a paid 256K OpenRouter route.