[Integrations](https://spiral.ai/integrations) › [Devin](https://spiral.ai/integrations/devin) › [Llama 4 Maverick](https://spiral.ai/integrations/devin/llama-4-maverick)

# How to Use Llama 4 Maverick in Devin

Llama 4 Maverick handles coding, multilingual chat, image questions, and creative writing. Compared with Llama 4 Scout, its documented context window is smaller: 1 million versus 10 million tokens.

|  |  |
| --- | --- |
| Model | Llama 4 Maverick |
| Available through | OpenRouter |
| Category | LLM |
| Price | Up to $0.005342 per run at the default inputs. $0.455 per 1M input tokens, $1.30 per 1M output tokens. |
| Returns | text, json |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with Llama 4 Maverick in Devin

- Use Llama 4 Maverick to review the current diff for behavior changes and likely bugs. For each finding, cite the file and relevant code, explain a concrete failure case, and suggest a focused test.
- Use Llama 4 Maverick to investigate the latest failing test using the test output and relevant project files. Explain the likely cause, propose the smallest fix, and identify what to rerun.
- Use Llama 4 Maverick to trace how an incoming request reaches the database in this project. Cite the functions involved and flag any step you cannot establish from the files.
- Use Llama 4 Maverick to draft Spanish and French versions of the English help text in this project. Preserve interpolation tokens and product names, and flag phrases that need more context.
- Use Llama 4 Maverick to draft release notes from the current diff and project documentation. Separate user-visible changes from internal changes, keep the draft under 400 words, and flag uncertain claims.

## How to set up Llama 4 Maverick in Devin

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

### Use Spiral's MCP server

1. **Add a custom MCP**: Open Customize › MCPs. Click + and choose New MCP. Fill in:

2. **Sign in to Spiral and approve**: In Customize › MCPs, open Spiral and choose Connect. Spiral opens: sign in, pick the organization and click Authorize. No API key.

3. **Check it**: On Spiral's configuration page, click Test tools. Devin lists Spiral's tools.

### Use the Spiral skill

1. **Send Devin 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 Llama 4 Maverick**: 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-llama/llama-4-maverick",
    "input": {
      "messages": [
        {
          "role": "system"
        }
      ]
    }
  }'
```

Then paste one of the prompts above: Devin finds Llama 4 Maverick on Spiral, shows the price and runs it.

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

## Llama 4 Maverick tools and prices

- [Meta: Llama 4 Maverick](https://spiral.ai/tools/openrouter/meta-llama/llama-4-maverick) (OpenRouter): Up to $0.005342 per run at the default inputs. $0.455 per 1M input tokens, $1.30 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 | — |  |
| logprobs |  | — |  |

## What can you do with Llama 4 Maverick in Devin?

Llama 4 Maverick can help turn project code and failure logs into review findings, explanations, and proposed fixes. Give it the relevant files and ask for evidence tied to the code before testing its suggestions.

### Review a code change

Use Llama 4 Maverick to examine a proposed change, explain likely defects, and suggest tests tied to specific behavior.

### Explain a failing test

Give Llama 4 Maverick the relevant code and failure output. Ask for a likely cause, a minimal correction, and a way to verify the fix.

### Draft multilingual support replies

Ask Llama 4 Maverick to draft replies in a supported language using supplied product guidance. Require it to flag questions the guidance does not answer.

### Summarize several documents

Have Llama 4 Maverick combine supplied documents into a brief that separates shared conclusions, disagreements, and unanswered questions.

### Describe product photos

Ask Llama 4 Maverick for English descriptions grounded in visible details. Require it to leave out unverified materials, dimensions, and performance claims.

### Answer questions about a chart

Give Llama 4 Maverick a chart image and ask in English about the visible trends. Request the labels or values supporting each observation.

### Explore different writing approaches

Use Llama 4 Maverick to draft several openings, explanations, or short stories from the same brief. Specify how the approaches should differ.

### Prepare answers from reference material

Ask Llama 4 Maverick to answer customer or internal questions using supplied reference text, with a supporting excerpt and a clear indication of missing information.

## Overview

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

## Llama 4 Maverick in other agents

- [Amp](https://spiral.ai/integrations/amp/llama-4-maverick)
- [ChatGPT](https://spiral.ai/integrations/chatgpt/llama-4-maverick)
- [Claude](https://spiral.ai/integrations/claude-desktop/llama-4-maverick)
- [Claude Code](https://spiral.ai/integrations/claude-code/llama-4-maverick)
- [Cline](https://spiral.ai/integrations/cline/llama-4-maverick)
- [Codex](https://spiral.ai/integrations/codex/llama-4-maverick)
- [Cursor](https://spiral.ai/integrations/cursor/llama-4-maverick)
- [dots](https://spiral.ai/integrations/dots/llama-4-maverick)
- [Factory Droid](https://spiral.ai/integrations/factory-droid/llama-4-maverick)
- [Gemini](https://spiral.ai/integrations/gemini/llama-4-maverick)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/llama-4-maverick)
- [Google Antigravity](https://spiral.ai/integrations/antigravity/llama-4-maverick)
- [goose](https://spiral.ai/integrations/goose/llama-4-maverick)
- [Grok](https://spiral.ai/integrations/grok/llama-4-maverick)
- [Grok Build](https://spiral.ai/integrations/grok-build/llama-4-maverick)
- [Hermes](https://spiral.ai/integrations/hermes/llama-4-maverick)
- [Instinct](https://spiral.ai/integrations/instinct/llama-4-maverick)
- [Manus](https://spiral.ai/integrations/manus/llama-4-maverick)
- [Muse](https://spiral.ai/integrations/muse/llama-4-maverick)
- [Muse Code](https://spiral.ai/integrations/muse-code/llama-4-maverick)
- [n8n](https://spiral.ai/integrations/n8n/llama-4-maverick)
- [Notion](https://spiral.ai/integrations/notion/llama-4-maverick)
- [OpenClaw](https://spiral.ai/integrations/openclaw/llama-4-maverick)
- [opencode](https://spiral.ai/integrations/opencode/llama-4-maverick)
- [Perplexity](https://spiral.ai/integrations/perplexity/llama-4-maverick)
- [Raycast](https://spiral.ai/integrations/raycast/llama-4-maverick)
- [VS Code](https://spiral.ai/integrations/vscode/llama-4-maverick)
- [Windsurf](https://spiral.ai/integrations/windsurf/llama-4-maverick)
- [Zapier](https://spiral.ai/integrations/zapier/llama-4-maverick)

## More Devin guides

- [Llama 3.3 70B Instruct](https://spiral.ai/integrations/devin/llama-3-3-70b-instruct)
- [Llama 4 Scout](https://spiral.ai/integrations/devin/llama-4-scout)
- [Llama 3.1 Euryale 70B v2.2](https://spiral.ai/integrations/devin/llama-3-1-euryale-70b-v2-2)
- [Llama Guard 4 12B](https://spiral.ai/integrations/devin/llama-guard-4-12b)
- [Llama 3.1 8B Instruct](https://spiral.ai/integrations/devin/llama-3-1-8b-instruct)
- [LongCat 2.0](https://spiral.ai/integrations/devin/longcat-2-0)

## Questions about Llama 4 Maverick in Devin

**Can Devin use Llama 4 Maverick?**

Yes. Add Spiral to Devin as an MCP server, and Devin can find and run Llama 4 Maverick like any built-in tool.

**How much does Llama 4 Maverick cost in Devin?**

Up to $0.005342 per run at the default inputs. $0.455 per 1M input tokens, $1.30 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Devin plan is separate.

**Do I need an OpenRouter account to use Llama 4 Maverick in Devin?**

No. Spiral runs Llama 4 Maverick on its own OpenRouter account. You sign in to Spiral once and never paste a key.

**How long does it take to set up Llama 4 Maverick in Devin?**

About two minutes: add a custom MCP, then sign in to Spiral and approve, then check it.

**Does Devin wait while Llama 4 Maverick runs?**

Llama 4 Maverick runs in the background. Devin starts the run, checks on it and gives you the text, json when it's ready.

**What is Llama 4 Maverick good at?**

Llama 4 Maverick is useful for code analysis, multilingual conversations, document summaries, image questions, and creative drafts. Give it the source material and a specific deliverable, such as a bug explanation or a summary with supporting excerpts.

**Which languages does Llama 4 Maverick support?**

Text support covers Arabic, English, French, German, Hindi, Indonesian, Italian, Portuguese, Spanish, Tagalog, Thai, and Vietnamese. Use English for image questions; documented image understanding is limited to English.

**How much context can Llama 4 Maverick handle?**

Its documented model context length is 1 million tokens, compared with 10 million for Llama 4 Scout. Treat that as context capacity, not a promise about the length of one answer.

**How should I prompt Llama 4 Maverick?**

State the task, supply the relevant material, and specify the audience, format, and length. Include an example when the output needs to follow a repeated pattern.

**Which setting matters for creative writing?**

Temperature is useful when you want more creative alternatives. Increase it while comparing drafts, and keep the audience, length, and factual requirements explicit.

