[Integrations](https://spiral.ai/integrations) › [Windsurf](https://spiral.ai/integrations/windsurf) › [Nemotron 3 Super](https://spiral.ai/integrations/windsurf/nemotron-3-super)

# How to Use Nemotron 3 Super in Windsurf

Use Nemotron 3 Super for complex code fixes, document analysis, and tasks that require several tool calls. Unlike Nemotron 3 Nano, it targets coding tasks that need deeper understanding of the codebase.

|  |  |
| --- | --- |
| Model | Nemotron 3 Super |
| Available through | OpenRouter |
| Category | LLM |
| Price | Up to $0.002135 per run at the default inputs. $0.0975 per 1M input tokens, $0.52 per 1M output tokens. |
| Returns | text, json |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with Nemotron 3 Super in Windsurf

- Use Nemotron 3 Super to investigate the failing tests in this project. Trace the failure through the relevant files, implement the smallest justified fix, and run a regression check.
- Use Nemotron 3 Super to review this repository's authentication flow. Map the relevant files, identify inconsistent permission checks, and support each finding with a code reference.
- Use Nemotron 3 Super to compare the migration guide with this project's current implementation. Produce an ordered change plan with affected files and an acceptance check for each stage.
- Use Nemotron 3 Super to trace how this project handles a request from entry point to database response. Explain where errors are transformed or lost, citing the relevant functions.
- Use Nemotron 3 Super to review the current diff against the project's requirements. Check callers and tests for behavior changes, then list findings by impact with supporting file references.

## How to set up Nemotron 3 Super in Windsurf

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

### Use Spiral's MCP server

1. **Add the server**: Open the Customize page from the Agents window and click MCPs. Under This computer, open the config file, or edit it directly: `~/.config/devin/mcp_config.json` on macOS and Linux, `%APPDATA%\devin\mcp_config.json` on Windows. Add this:

```json
{
  "mcpServers": {
    "spiral": {
      "url": "https://api.spiral.ai/mcp"
    }
  }
}
```

2. **Sign in**: Click Authenticate on the Spiral server in Customize › MCPs, or let the agent call Spiral once: your browser opens. Sign in to Spiral, pick the organization and click Authorize. No API key.

```bash
devin mcp login spiral
```

3. **Check it**: Spiral is listed under This computer in Customize › MCPs. From the terminal:

```bash
devin mcp list
```

### Use the Spiral skill

1. **Send Windsurf 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 Super**: 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-super-120b-a12b",
    "input": {
      "messages": [
        {
          "role": "system"
        }
      ]
    }
  }'
```

Then paste one of the prompts above: Windsurf finds Nemotron 3 Super on Spiral, shows the price and runs it.

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

## Nemotron 3 Super tools and prices

- [NVIDIA: Nemotron 3 Super](https://spiral.ai/tools/openrouter/nvidia/nemotron-3-super-120b-a12b) (OpenRouter): Up to $0.002135 per run at the default inputs. $0.0975 per 1M input tokens, $0.52 per 1M output tokens.

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| seed |  | — |  |
| stop |  | — |  |
| min_p | 0 to 1 | — |  |
| tools |  | — |  |
| top_k | 0 to … | — |  |
| top_p | 0 to 1 | — |  |

## What can you do with Nemotron 3 Super in Windsurf?

Trace bugs and plan changes across a codebase with Nemotron 3 Super. Give the agent project context and acceptance checks so it can connect code changes to the behavior you need.

### Fix bugs across several files

Trace a reported failure through the codebase, reproduce it, and propose a patch with a regression check.

### Plan a codebase migration

Compare existing code with a migration specification. Identify affected components, dependencies, and checks for each stage.

### Compare long reports

Analyze supplied reports for changes, contradictions, and unanswered questions. Request a source passage for each finding.

### Answer questions from retrieved documentation

Use retrieved passages to answer a specific question, cite the evidence, and identify anything the passages do not establish.

### Triage IT tickets

Compare incoming tickets with supplied routing rules and troubleshooting guidance. Suggest a queue, next step, and missing information.

### Draft multilingual support replies

Draft replies from approved guidance in English, French, German, Italian, Japanese, Spanish, or Chinese. Preserve product names and technical terms.

### Coordinate tasks across tools

Break a request into tool actions with explicit success criteria. Use returned evidence to decide the next action.

### Compare financial disclosures

Read supplied financial reports to identify changed assumptions and explanations. Keep reported figures separate from inferred conclusions.

## Overview

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

## Nemotron 3 Super in other agents

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

## More Windsurf guides

- [Nemotron 3 Nano 30B A3B](https://spiral.ai/integrations/windsurf/nemotron-3-nano-30b-a3b)
- [Nemotron 3 Ultra](https://spiral.ai/integrations/windsurf/nemotron-3-ultra)
- [MythoMax 13B](https://spiral.ai/integrations/windsurf/mythomax-13b)
- [Nemotron 3.5 Content Safety](https://spiral.ai/integrations/windsurf/nemotron-3-5-content-safety)
- [Muse Wonder](https://spiral.ai/integrations/windsurf/muse-wonder)
- [Nemotron 3.5 Lightning](https://spiral.ai/integrations/windsurf/nemotron-3-5-lightning)

## Questions about Nemotron 3 Super in Windsurf

**Can Windsurf use Nemotron 3 Super?**

Yes. Add Spiral to Windsurf as an MCP server, and Windsurf can find and run Nemotron 3 Super like any built-in tool.

**How much does Nemotron 3 Super cost in Windsurf?**

Up to $0.002135 per run at the default inputs. $0.0975 per 1M input tokens, $0.52 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Windsurf plan is separate.

**Do I need an OpenRouter account to use Nemotron 3 Super in Windsurf?**

No. Spiral runs Nemotron 3 Super 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 Super in Windsurf?**

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

**Does Windsurf wait while Nemotron 3 Super runs?**

Nemotron 3 Super runs in the background. Windsurf starts the run, checks on it and gives you the text, json when it's ready.

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

Complex coding tasks, analysis across long documents, and workflows that require several tool calls. Give it a concrete goal and a way to check whether the result meets that goal.

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

NVIDIA specifies up to 1 million tokens of context. That describes the context window, not a promise of a million-token answer.

**Can Nemotron 3 Super miss details in long documents?**

Yes. NVIDIA's RULER score falls from 95.22 at 512,000 tokens to 91.64 at 1 million tokens, so ask for supporting passages and check important conclusions against them.

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

English, French, German, Italian, Japanese, Spanish, and Chinese. Specify the response language and any terms that should stay untranslated.

**Which sampling settings should I start with?**

NVIDIA recommends temperature 1.0 and top-p 0.95 across tasks. Start there and compare results on representative examples before changing them.

