How to Use Llama 4 Scout in Devin
Llama 4 Scout suits summaries across multiple documents and questions about large codebases. Its distinguishing feature is Meta's documented 10-million-token input context window.
Prompts to try with Llama 4 Scout in Devin
Copy an example into Devin, or ask in your own words.
Use Llama 4 Scout to trace how an incoming request reaches the database in this project. Cite file paths and function names, and mark any missing links.
Use Llama 4 Scout to compare the project's architecture notes with its implementation. List contradictions with a document reference and a code excerpt for each.
Use Llama 4 Scout to review the source files involved in authentication. Explain the control flow and identify questions that require runtime evidence.
Use Llama 4 Scout to summarize the migration notes and affected modules in this repository. Return dependencies, likely edit locations, and unresolved assumptions.
Use Llama 4 Scout at a low temperature to extract configuration names and their documented purposes from this project's files. Cite each source and flag conflicting definitions.
How to set up Llama 4 Scout in Devin
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Add a custom MCP
Open Customize › MCPs. Click + and choose New MCP. Fill in:
- 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.
- Check it
On Spiral's configuration page, click Test tools. Devin lists Spiral's tools.
Use the Spiral skill
- Send Devin this message
Your agent reads Spiral's skill and sets itself up.
set up https://spiral.ai/SKILL.md
- Sign in or paste a key
Approve Spiral in the browser, or give the agent a key from the console.
Use the Spiral CLI
- Install and sign in
Installs the Spiral CLI, signs you in and adds its skill to your coding agents.
npx -y spiral-cli@latest init
Call the Spiral API
- Create an API key
In the console, under API keys.
- Run Llama 4 Scout
One request; the result comes back by polling or a webhook.
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-scout", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Devin finds Llama 4 Scout on Spiral, shows the price and runs it.
Llama 4 Scout providers and prices
Spiral runs Llama 4 Scout through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.001603 per run at the default inputs. $0.13 per 1M input tokens, $0.39 per 1M output tokens.
Settings you can ask for
Ask Devin for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Seed | a whole number | ||
| Min p | 0 to 1 | ||
| Top k | a whole number from 0 | ||
| Top p | 0 to 1 | ||
| Max tokens | 1 to 16384 | The most output tokens to generate. | |
| Temperature | 0 to 2 | ||
| Presence penalty | -2 to 2 | ||
| Frequency penalty | -2 to 2 | ||
| Repetition penalty | 0 to 2 |
Every API field of Meta: Llama 4 Scout is on its tool page: open its card above.
What can you do with Llama 4 Scout in Devin?
Llama 4 Scout can help trace behavior across project files and summarize the documentation behind a change. Ask for file references and supporting code excerpts so you can inspect each conclusion.
Compare research reports
Ask Llama 4 Scout to summarize several reports, separate agreements from contradictions, and attach a source label to each finding.
Trace behavior across a codebase
Give Llama 4 Scout the relevant source files and ask it to follow a request across modules, citing functions and paths.
Review a long support history
Have Llama 4 Scout organize supplied conversations into recurring problems, attempted fixes, and unresolved questions.
Prepare multilingual summaries
Ask Llama 4 Scout to summarize supplied material in Spanish, French, or another supported text language while preserving names and technical terms.
Compare product photos
Give Llama 4 Scout two product images and request a written list of visible differences, separating observations from guesses.
Build an activity-based briefing
Use Llama 4 Scout to turn supplied activity records into a chronology of completed work, repeated requests, and open commitments.
Draft image descriptions
Ask Llama 4 Scout for concise descriptions of supplied images, with explicit uncertainty where a detail is unclear.
More on Llama 4 Scout and Devin
Overview
Llama 4 Scout in other agents
Questions about Llama 4 Scout in Devin
Can Devin use Llama 4 Scout?
Yes. Add Spiral to Devin as an MCP server, and Devin can find and run Llama 4 Scout like any built-in tool.
How much does Llama 4 Scout cost in Devin?
Up to $0.001603 per run at the default inputs. $0.13 per 1M input tokens, $0.39 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 Scout in Devin?
No. Spiral runs Llama 4 Scout 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 Scout 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 Scout runs?
Llama 4 Scout 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 Scout useful for?
Start with summaries across multiple documents, questions about large codebases, or analysis of long activity histories. Give each source a label and ask Scout to identify the evidence behind its conclusions.
What does Llama 4 Scout's 10-million-token context mean?
Meta documents it as an input context length. Treat it as capacity for source material, rather than a promise of a 10-million-token answer.
Which languages does Llama 4 Scout support?
The 12 supported text languages are Arabic, English, French, German, Hindi, Indonesian, Italian, Portuguese, Spanish, Tagalog, Thai, and Vietnamese. Specify the language you want in the answer.
Can Llama 4 Scout compare images?
Yes, it can analyze image inputs and return a written comparison. Meta tested image understanding with up to five input images; that is a tested scope, not proof of equal performance with larger sets.
Can I trust every detail in a Llama 4 Scout answer?
No. Meta warns that it can produce inaccurate responses, so ask for supporting passages and check consequential claims against the original material.
Start using Llama 4 Scout in Devin
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Devin:
set up https://spiral.ai/SKILL.md


