[Integrations](https://spiral.ai/integrations) › [opencode](https://spiral.ai/integrations/opencode) › [Nemotron 3 Nano 30B A3B](https://spiral.ai/integrations/opencode/nemotron-3-nano-30b-a3b)

# How to Use Nemotron 3 Nano 30B A3B in opencode

Nemotron 3 Nano 30B A3B handles debugging, document summaries, and tool-driven tasks. Switch between reasoning for harder problems and direct answers for routine requests.

|  |  |
| --- | --- |
| Model | Nemotron 3 Nano 30B A3B |
| Available through | OpenRouter |
| Category | LLM |
| Price | Up to $0.001282 per run at the default inputs. $0.078 per 1M input tokens, $0.312 per 1M output tokens. |
| Returns | text, json |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with Nemotron 3 Nano 30B A3B in opencode

- Use Nemotron 3 Nano 30B A3B to inspect the failing tests and relevant source files. Reproduce one failure, propose the smallest fix, and run the affected tests.
- Use Nemotron 3 Nano 30B A3B to trace authentication through this project's code. Cite file locations and identify where missing or expired credentials are handled.
- Use Nemotron 3 Nano 30B A3B to compare the implementation with the project's documented requirements. Report discrepancies with code references and a proposed test for each.
- Use Nemotron 3 Nano 30B A3B to review the changed parsing code. Draft tests for empty input, malformed records, and duplicate values, then run them.
- Use Nemotron 3 Nano 30B A3B to analyze the calculation in this project with reasoning enabled. Check units, rounding, and boundary conditions before proposing changes.

## How to set up Nemotron 3 Nano 30B A3B in opencode

Pick the way that suits you. Each takes about two minutes.

### Use Spiral's MCP server

1. **Add the server**: Leave out `--global` to add Spiral to the current project only. No header and no client ID or secret: opencode registers itself with Spiral.

```bash
opencode mcp add spiral --global --url https://api.spiral.ai/mcp
```

2. **Sign in**: Your browser opens. Sign in to Spiral, pick the organization and click Authorize, then return to the terminal. No API key. Inside opencode, `/mcps` starts the same sign-in.

```bash
opencode mcp auth spiral
```

3. **Check it**: Spiral shows as connected.

```bash
opencode mcp list
```

### Use the Spiral skill

1. **Send opencode this message**: Your agent reads Spiral's skill and sets itself up.

```
set up https://spiral.ai/SKILL.md
```

2. **Sign in or paste a key**: Approve Spiral in the browser, or give the agent a key from the console.

### Use the Spiral CLI

1. **Install and sign in**: Installs the Spiral CLI, signs you in and adds its skill to your coding agents.

```bash
npx -y spiral-cli@latest init
```

### Call the Spiral API

1. **Create an API key**: In the console, under API keys.

2. **Run Nemotron 3 Nano 30B A3B**: One request; the result comes back by polling or a webhook.

```bash
curl https://api.spiral.ai/v1/runs \
  -H "Authorization: Bearer $SPIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "provider": "openrouter",
    "model": "nvidia/nemotron-3-nano-30b-a3b",
    "input": {
      "messages": [
        {
          "role": "system"
        }
      ]
    }
  }'
```

Then paste one of the prompts above: opencode finds Nemotron 3 Nano 30B A3B on Spiral, shows the price and runs it.

Setup guide in the docs: https://spiral.ai/docs/quickstarts/opencode

## Nemotron 3 Nano 30B A3B tools and prices

- [NVIDIA: Nemotron 3 Nano 30B A3B](https://spiral.ai/tools/openrouter/nvidia/nemotron-3-nano-30b-a3b) (OpenRouter): Up to $0.001282 per run at the default inputs. $0.078 per 1M input tokens, $0.312 per 1M output tokens.

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| seed |  | — |  |
| stop |  | — |  |
| tools |  | — |  |
| top_k | 0 to … | — |  |
| top_p | 0 to 1 | — |  |
| reasoning |  | — | Reasoning effort; reasoning tokens count toward max_tokens. |

## What can you do with Nemotron 3 Nano 30B A3B in opencode?

Turn failing tests and project code into focused repair proposals with Nemotron 3 Nano. Use it to trace dependencies, explain failures, and draft patches for testing.

### Repair a reproducible bug

Give Nemotron 3 Nano a failing test and relevant code. Request a focused patch, then test the proposed fix.

### Summarize lengthy discussions

Turn meeting transcripts or support conversations into decisions, unresolved questions, and action items, with supporting excerpts.

### Answer from reference text

Use retrieved documentation to draft answers that cite passages and identify questions the supplied material cannot resolve.

### Extract records into JSON

Convert order notes or support messages into a defined JSON structure. Specify required values and how to represent missing details.

### Draft multilingual support replies

Prepare replies in English, German, Spanish, French, Italian, or Japanese using supplied policies and customer messages.

### Work through a tool sequence

Combine lookup and calculation steps to answer an operational question. Require the final answer to distinguish retrieved evidence from calculations.

### Check mathematical assumptions

Work through a quantitative problem, state assumptions, and check the result against its units and constraints.

## Overview

- [OpenRouter](https://spiral.ai/providers/openrouter)
- [All opencode guides](https://spiral.ai/integrations/opencode)

## Nemotron 3 Nano 30B A3B in other agents

- [Amp](https://spiral.ai/integrations/amp/nemotron-3-nano-30b-a3b)
- [ChatGPT](https://spiral.ai/integrations/chatgpt/nemotron-3-nano-30b-a3b)
- [Claude](https://spiral.ai/integrations/claude-desktop/nemotron-3-nano-30b-a3b)
- [Claude Code](https://spiral.ai/integrations/claude-code/nemotron-3-nano-30b-a3b)
- [Cline](https://spiral.ai/integrations/cline/nemotron-3-nano-30b-a3b)
- [Codex](https://spiral.ai/integrations/codex/nemotron-3-nano-30b-a3b)
- [Cursor](https://spiral.ai/integrations/cursor/nemotron-3-nano-30b-a3b)
- [Devin](https://spiral.ai/integrations/devin/nemotron-3-nano-30b-a3b)
- [dots](https://spiral.ai/integrations/dots/nemotron-3-nano-30b-a3b)
- [Factory Droid](https://spiral.ai/integrations/factory-droid/nemotron-3-nano-30b-a3b)
- [Gemini](https://spiral.ai/integrations/gemini/nemotron-3-nano-30b-a3b)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/nemotron-3-nano-30b-a3b)
- [Google Antigravity](https://spiral.ai/integrations/antigravity/nemotron-3-nano-30b-a3b)
- [goose](https://spiral.ai/integrations/goose/nemotron-3-nano-30b-a3b)
- [Grok](https://spiral.ai/integrations/grok/nemotron-3-nano-30b-a3b)
- [Grok Build](https://spiral.ai/integrations/grok-build/nemotron-3-nano-30b-a3b)
- [Hermes](https://spiral.ai/integrations/hermes/nemotron-3-nano-30b-a3b)
- [Instinct](https://spiral.ai/integrations/instinct/nemotron-3-nano-30b-a3b)
- [Manus](https://spiral.ai/integrations/manus/nemotron-3-nano-30b-a3b)
- [Muse](https://spiral.ai/integrations/muse/nemotron-3-nano-30b-a3b)
- [Muse Code](https://spiral.ai/integrations/muse-code/nemotron-3-nano-30b-a3b)
- [n8n](https://spiral.ai/integrations/n8n/nemotron-3-nano-30b-a3b)
- [Notion](https://spiral.ai/integrations/notion/nemotron-3-nano-30b-a3b)
- [OpenClaw](https://spiral.ai/integrations/openclaw/nemotron-3-nano-30b-a3b)
- [Perplexity](https://spiral.ai/integrations/perplexity/nemotron-3-nano-30b-a3b)
- [Raycast](https://spiral.ai/integrations/raycast/nemotron-3-nano-30b-a3b)
- [VS Code](https://spiral.ai/integrations/vscode/nemotron-3-nano-30b-a3b)
- [Windsurf](https://spiral.ai/integrations/windsurf/nemotron-3-nano-30b-a3b)
- [Zapier](https://spiral.ai/integrations/zapier/nemotron-3-nano-30b-a3b)

## More opencode guides

- [MythoMax 13B](https://spiral.ai/integrations/opencode/mythomax-13b)
- [Nemotron 3 Super](https://spiral.ai/integrations/opencode/nemotron-3-super)
- [Muse Wonder](https://spiral.ai/integrations/opencode/muse-wonder)
- [Nemotron 3 Ultra](https://spiral.ai/integrations/opencode/nemotron-3-ultra)
- [Muse Spark Open](https://spiral.ai/integrations/opencode/muse-spark-open)
- [Nemotron 3.5 Content Safety](https://spiral.ai/integrations/opencode/nemotron-3-5-content-safety)

## Questions about Nemotron 3 Nano 30B A3B in opencode

**Can opencode use Nemotron 3 Nano 30B A3B?**

Yes. Add Spiral to opencode as an MCP server, and opencode can find and run Nemotron 3 Nano 30B A3B like any built-in tool.

**How much does Nemotron 3 Nano 30B A3B cost in opencode?**

Up to $0.001282 per run at the default inputs. $0.078 per 1M input tokens, $0.312 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your opencode plan is separate.

**Do I need an OpenRouter account to use Nemotron 3 Nano 30B A3B in opencode?**

No. Spiral runs Nemotron 3 Nano 30B A3B on its own OpenRouter account. You sign in to Spiral once and never paste a key.

**How long does it take to set up Nemotron 3 Nano 30B A3B in opencode?**

About two minutes: add the server, then sign in, then check it.

**Does opencode wait while Nemotron 3 Nano 30B A3B runs?**

Nemotron 3 Nano 30B A3B runs in the background. opencode starts the run, checks on it and gives you the text, json when it's ready.

**What is Nemotron 3 Nano good at?**

Start with debugging, document summaries, and questions answered from supplied reference text. NVIDIA also targets assistant workflows and information retrieval.

**How much context can Nemotron 3 Nano handle?**

NVIDIA documents up to one million tokens at the model level. For long-document tasks, ask it to identify supporting passages alongside its conclusions.

**Which languages does Nemotron 3 Nano support?**

NVIDIA lists English, German, Spanish, French, Italian, and Japanese. Specify the response language and any terminology that should stay unchanged.

**Can Nemotron 3 Nano produce structured JSON?**

Yes. NVIDIA trained it specifically for JSON schema adherence; provide the required structure and define how missing information should appear.

**When should I use reasoning?**

Use reasoning for logical deductions, technical problems, and questions with several dependent steps. NVIDIA's guidance associates it with better accuracy on challenging questions.

