[Integrations](https://spiral.ai/integrations) › [Windsurf](https://spiral.ai/integrations/windsurf) › [Muse Glimmer 30B](https://spiral.ai/integrations/windsurf/muse-glimmer-30b)

# How to Use Muse Glimmer 30B in Windsurf

Muse Glimmer 30B suits coding, sequential tool use, and tasks that need recovery after a failed tool call. Its 30B design combines text and image understanding with adjustable reasoning effort.

|  |  |
| --- | --- |
| Model | Muse Glimmer 30B |
| Available through | OpenRouter |
| Category | LLM |
| Price | Up to $0.006404 per run at the default inputs. $0.39 per 1M input tokens, $1.56 per 1M output tokens. |
| Returns | text, json |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with Muse Glimmer 30B in Windsurf

- Use Muse Glimmer 30B to investigate the failing tests in this project. Trace the first failure, make a focused fix, and report which checks passed.
- Use Muse Glimmer 30B to review the current diff for correctness bugs. Return at most five findings, each with a file location, failure scenario, and suggested fix.
- Use Muse Glimmer 30B to inspect this project's request retry logic. Identify where a temporary tool or network failure could halt the workflow, then propose a bounded recovery strategy.
- Use Muse Glimmer 30B to compare the screenshot in this project with the relevant UI code. Identify the likely cause of the visible error and verify any proposed change.
- Use Muse Glimmer 30B to evaluate the saved assistant responses against the project's rubric. Explain each score with evidence from the response and flag cases needing human review.

## How to set up Muse Glimmer 30B 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 Muse Glimmer 30B**: 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": "meta/muse-glimmer-30b",
    "input": {
      "messages": [
        {
          "role": "system"
        }
      ]
    }
  }'
```

Then paste one of the prompts above: Windsurf finds Muse Glimmer 30B on Spiral, shows the price and runs it.

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

## Muse Glimmer 30B tools and prices

- [Meta: Muse Glimmer 30B](https://spiral.ai/tools/openrouter/meta/muse-glimmer-30b) (OpenRouter): Up to $0.008005 per run at the default inputs. $0.455 per 1M input tokens, $1.95 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 Muse Glimmer 30B in Windsurf?

Use Muse Glimmer to investigate bugs, review changes, and work through coding tasks that require several actions. Give it a concrete success condition and ask it to check the result against that condition.

### Debug a failing application

Trace a failure through the relevant code, propose a focused fix, and check the result against the original error.

### Review a code change

Ask for concrete bugs, their consequences, and suggested fixes, with a strict limit on speculative findings.

### Investigate a question with tools

Gather evidence through successive searches, resolve conflicting results, and produce an answer tied to the retrieved sources.

### Recover a stalled workflow

Inspect an unsuccessful tool result, identify the likely cause, and choose a revised next action.

### Explain a chart

Turn a chart image into a short explanation of the visible trend, labels, and exceptions.

### Extract information from a scan

Read a scanned document and organize the requested details into a concise text report, marking anything unclear.

### Diagnose a confusing screen

Inspect a screenshot, explain the visible state, and suggest the next troubleshooting check.

### Evaluate competing answers

Compare candidate responses against an explicit rubric for correctness, completeness, and instruction following.

## Overview

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

## Muse Glimmer 30B in other agents

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

## More Windsurf guides

- [Muse Ember](https://spiral.ai/integrations/windsurf/muse-ember)
- [Muse Glimmer 30B (batch)](https://spiral.ai/integrations/windsurf/muse-glimmer-30b-batch)
- [Motif 3](https://spiral.ai/integrations/windsurf/motif-3)
- [Muse Spark 1.1](https://spiral.ai/integrations/windsurf/muse-spark-1-1)
- [Morph V3 Large (batch)](https://spiral.ai/integrations/windsurf/morph-v3-large-batch)
- [Muse Spark 1.2](https://spiral.ai/integrations/windsurf/muse-spark-1-2)

## Questions about Muse Glimmer 30B in Windsurf

**Can Windsurf use Muse Glimmer 30B?**

Yes. Add Spiral to Windsurf as an MCP server, and Windsurf can find and run Muse Glimmer 30B like any built-in tool.

**How much does Muse Glimmer 30B cost in Windsurf?**

Up to $0.006404 per run at the default inputs. $0.39 per 1M input tokens, $1.56 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 Muse Glimmer 30B in Windsurf?**

No. Spiral runs Muse Glimmer 30B on its own OpenRouter account. You sign in to Spiral once and never paste a key.

**How long does it take to set up Muse Glimmer 30B in Windsurf?**

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

**Does Windsurf wait while Muse Glimmer 30B runs?**

Muse Glimmer 30B runs in the background. Windsurf starts the run, checks on it and gives you the text, json when it's ready.

**What is Muse Glimmer best suited for?**

Use Muse Glimmer for coding, tasks involving several tool calls, and workflows that need to recover from tool errors. Meta also identifies evaluating other models' answers as an intended use.

**Can Muse Glimmer understand images or generate them?**

Muse Glimmer can interpret images alongside text, including screenshots, charts, and scanned documents. It produces text, so use it to explain or extract information from an image.

**How much context can Muse Glimmer handle?**

Meta documents a default context window of 128K tokens and support for longer contexts. Treat 128K as the documented default rather than assuming an unlimited conversation history.

**Does Muse Glimmer work across languages?**

Muse Glimmer was trained on data from more than 100 languages. Test representative examples in your target language before relying on it for translation or customer-facing writing.

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

Start with the documented recommendation: temperature 1.0, top-p 0.95, and top-k 64. Compare results on a fixed set of your own tasks before changing those settings.

