[Integrations](https://spiral.ai/integrations) › [Devin](https://spiral.ai/integrations/devin) › [Mistral Small 3.1 24B Instruct](https://spiral.ai/integrations/devin/mistral-small-3-1-24b-instruct)

# How to Use Mistral Small 3.1 24B Instruct in Devin

Mistral Small 3.1 24B Instruct suits document summaries, coding help, and tool-driven assistants. Compared with Mistral Small 3, it adds image understanding and extends context to 128k tokens.

|  |  |
| --- | --- |
| Model | Mistral Small 3.1 24B Instruct |
| Available through | OpenRouter |
| Category | LLM |
| Price | Up to $0.002972 per run at the default inputs. $0.4563 per 1M input tokens, $0.7215 per 1M output tokens. |
| Returns | text, json |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with Mistral Small 3.1 24B Instruct in Devin

- Use Mistral Small 3.1 24B Instruct to review the current diff against the requirements in this repository. Report likely defects with the file, function, triggering input, and a suggested test.
- Use Mistral Small 3.1 24B Instruct at temperature 0.15 to extract requirements from the project's design documents. Return JSON with requirement, source path, supporting excerpt, and unresolved question.
- Use Mistral Small 3.1 24B Instruct to inspect the parser and its existing tests. Propose cases for empty input, malformed input, and boundary values, explaining the expected behavior from the code.
- Use Mistral Small 3.1 24B Instruct to draft French and German versions of the English interface strings in this project. Preserve interpolation tokens and product names, and list phrases that need human review.
- Use Mistral Small 3.1 24B Instruct to trace how a request reaches the database in this repository. Give an ordered list of functions and identify where validation and error handling occur.

## How to set up Mistral Small 3.1 24B Instruct 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 Mistral Small 3.1 24B Instruct**: 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": "mistralai/mistral-small-3.1-24b-instruct",
    "input": {
      "messages": [
        {
          "role": "system"
        }
      ]
    }
  }'
```

Then paste one of the prompts above: Devin finds Mistral Small 3.1 24B Instruct on Spiral, shows the price and runs it.

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

## Mistral Small 3.1 24B Instruct tools and prices

- [Mistral: Mistral Small 3.1 24B](https://spiral.ai/tools/openrouter/mistralai/mistral-small-3.1-24b-instruct) (OpenRouter): Up to $0.002972 per run at the default inputs. $0.4563 per 1M input tokens, $0.7215 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 Mistral Small 3.1 24B Instruct in Devin?

Turn project code and technical documents into review notes, test ideas, and implementation explanations with Mistral Small 3.1. Give the agent a bounded task and a concrete output format so you can check the result against the project.

### Technical document briefs

Turn a technical specification into requirements, unresolved questions, and decisions. Ask for a supporting excerpt beside each finding.

### Code review preparation

Review a proposed patch against its requirements and identify edge cases to test. Request a short explanation tied to the relevant function.

### Structured request extraction

Convert customer messages into JSON containing the request, relevant dates, and missing information. Require null for details the message does not provide.

### Multilingual support drafts

Draft responses in English, French, German, or Japanese using supplied support guidance. Preserve product names and flag questions the guidance cannot answer.

### Daily operations summaries

Summarize incoming status updates into blockers, decisions, and next actions. Carry over the original record IDs so each item can be checked.

### Tool-assisted support lookup

Use function calling to request relevant support records, then draft an answer from the returned information. Ask for clarification when the lookup needs missing details.

### Math solution review

Review a supplied calculation, explain the method, and identify assumptions or inconsistent units. Use a calculator or tests to verify the final numbers.

## Overview

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

## Mistral Small 3.1 24B Instruct in other agents

- [Amp](https://spiral.ai/integrations/amp/mistral-small-3-1-24b-instruct)
- [ChatGPT](https://spiral.ai/integrations/chatgpt/mistral-small-3-1-24b-instruct)
- [Claude](https://spiral.ai/integrations/claude-desktop/mistral-small-3-1-24b-instruct)
- [Claude Code](https://spiral.ai/integrations/claude-code/mistral-small-3-1-24b-instruct)
- [Cline](https://spiral.ai/integrations/cline/mistral-small-3-1-24b-instruct)
- [Codex](https://spiral.ai/integrations/codex/mistral-small-3-1-24b-instruct)
- [Cursor](https://spiral.ai/integrations/cursor/mistral-small-3-1-24b-instruct)
- [dots](https://spiral.ai/integrations/dots/mistral-small-3-1-24b-instruct)
- [Factory Droid](https://spiral.ai/integrations/factory-droid/mistral-small-3-1-24b-instruct)
- [Gemini](https://spiral.ai/integrations/gemini/mistral-small-3-1-24b-instruct)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/mistral-small-3-1-24b-instruct)
- [Google Antigravity](https://spiral.ai/integrations/antigravity/mistral-small-3-1-24b-instruct)
- [goose](https://spiral.ai/integrations/goose/mistral-small-3-1-24b-instruct)
- [Grok](https://spiral.ai/integrations/grok/mistral-small-3-1-24b-instruct)
- [Grok Build](https://spiral.ai/integrations/grok-build/mistral-small-3-1-24b-instruct)
- [Hermes](https://spiral.ai/integrations/hermes/mistral-small-3-1-24b-instruct)
- [Instinct](https://spiral.ai/integrations/instinct/mistral-small-3-1-24b-instruct)
- [Manus](https://spiral.ai/integrations/manus/mistral-small-3-1-24b-instruct)
- [Muse](https://spiral.ai/integrations/muse/mistral-small-3-1-24b-instruct)
- [Muse Code](https://spiral.ai/integrations/muse-code/mistral-small-3-1-24b-instruct)
- [n8n](https://spiral.ai/integrations/n8n/mistral-small-3-1-24b-instruct)
- [Notion](https://spiral.ai/integrations/notion/mistral-small-3-1-24b-instruct)
- [OpenClaw](https://spiral.ai/integrations/openclaw/mistral-small-3-1-24b-instruct)
- [opencode](https://spiral.ai/integrations/opencode/mistral-small-3-1-24b-instruct)
- [Perplexity](https://spiral.ai/integrations/perplexity/mistral-small-3-1-24b-instruct)
- [Raycast](https://spiral.ai/integrations/raycast/mistral-small-3-1-24b-instruct)
- [VS Code](https://spiral.ai/integrations/vscode/mistral-small-3-1-24b-instruct)
- [Windsurf](https://spiral.ai/integrations/windsurf/mistral-small-3-1-24b-instruct)
- [Zapier](https://spiral.ai/integrations/zapier/mistral-small-3-1-24b-instruct)

## More Devin guides

- [Mistral Small 3](https://spiral.ai/integrations/devin/mistral-small-3)
- [Mistral Small 3.2 24B](https://spiral.ai/integrations/devin/mistral-small-3-2-24b)
- [Mistral Saba](https://spiral.ai/integrations/devin/mistral-saba)
- [Mistral Small 4](https://spiral.ai/integrations/devin/mistral-small-4)
- [Mistral NeMo](https://spiral.ai/integrations/devin/mistral-nemo)
- [Mixtral 8x22B Instruct](https://spiral.ai/integrations/devin/mixtral-8x22b-instruct)

## Questions about Mistral Small 3.1 24B Instruct in Devin

**Can Devin use Mistral Small 3.1 24B Instruct?**

Yes. Add Spiral to Devin as an MCP server, and Devin can find and run Mistral Small 3.1 24B Instruct like any built-in tool.

**How much does Mistral Small 3.1 24B Instruct cost in Devin?**

Up to $0.002972 per run at the default inputs. $0.4563 per 1M input tokens, $0.7215 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 Mistral Small 3.1 24B Instruct in Devin?**

No. Spiral runs Mistral Small 3.1 24B Instruct on its own OpenRouter account. You sign in to Spiral once and never paste a key.

**How long does it take to set up Mistral Small 3.1 24B Instruct in Devin?**

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

**Does Devin wait while Mistral Small 3.1 24B Instruct runs?**

Mistral Small 3.1 24B Instruct runs in the background. Devin starts the run, checks on it and gives you the text, json when it's ready.

**What is Mistral Small 3.1 good at?**

Use it for document summaries, coding assistance, mathematical reasoning, and conversational tasks. A useful starting task is turning a long technical document into decisions, open questions, and supporting excerpts.

**How much context can Mistral Small 3.1 handle?**

Mistral documents a context window of up to 128k tokens. Treat that as context capacity, rather than a promised answer length; ask for a focused result from the material you supply.

**Which languages does Mistral Small 3.1 support?**

Mistral lists dozens, including English, French, German, Greek, Hindi, Indonesian, Italian, and Japanese. Specify the response language and identify terms that should stay untranslated.

**What temperature should I use with Mistral Small 3.1?**

Mistral recommends a relatively low temperature, such as 0.15. Start there for extraction, classification, and other tasks with a defined answer format.

**Can Mistral Small 3.1 return JSON and call functions?**

Yes. Mistral lists native function calling and JSON output as capabilities; for extraction tasks, explicitly request JSON and describe the structure you want.

