[Integrations](https://spiral.ai/integrations) › [Perplexity](https://spiral.ai/integrations/perplexity) › [Mistral Small 4](https://spiral.ai/integrations/perplexity/mistral-small-4)

# How to Use Mistral Small 4 in Perplexity

Mistral Small 4 handles code exploration, document analysis, and multilingual chat with a 256k context window. Adjustable reasoning lets you reserve deeper analysis for difficult problems and use direct responses for routine work.

|  |  |
| --- | --- |
| Model | Mistral Small 4 |
| Available through | OpenRouter |
| Category | LLM |
| Price | Up to $0.003202 per run at the default inputs. $0.195 per 1M input tokens, $0.78 per 1M output tokens. |
| Returns | text, json |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with Mistral Small 4 in Perplexity

- Use Mistral Small 4 to compare the two policy excerpts in this conversation. List conflicting requirements, cite the relevant sections, and identify questions the excerpts leave unanswered.
- Use Mistral Small 4 with deeper reasoning to compare the three proposals described above. Evaluate each against my requirements and explain which missing facts could change the recommendation.
- Use Mistral Small 4 to draft a French reply to the customer message above using only the supplied policy. Preserve product names and list any questions the policy cannot answer.
- Use Mistral Small 4 to explain the chart image in this conversation. Describe the main trend, cite visible labels and values, and mark anything you cannot read.
- Use Mistral Small 4 with reasoning turned off to extract decisions, owners, and due dates from the meeting transcript above. Return a table and mark missing details as unspecified.

## How to set up Mistral Small 4 in Perplexity

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

### Use the Perplexity connector

1. **Add a remote connector**: In Perplexity, open Plugins in the sidebar (or Customize › Connectors if you don't see Plugins yet). Choose Add custom › Add MCP connector, then Remote. Name it Spiral and paste this MCP server URL:

```text
https://api.spiral.ai/mcp
```

2. **Sign in to Spiral and approve**: Select the Spiral card on the Plugins page to start sign-in. Spiral opens: sign in, pick the organization and click Authorize. No API key.

### Use the Spiral skill

1. **Send Perplexity 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 4**: 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-2603",
    "input": {
      "messages": [
        {
          "role": "system"
        }
      ]
    }
  }'
```

Then paste one of the prompts above: Perplexity finds Mistral Small 4 on Spiral, shows the price and runs it.

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

## Mistral Small 4 tools and prices

- [Mistral: Mistral Small 4](https://spiral.ai/tools/openrouter/mistralai/mistral-small-2603) (OpenRouter): Up to $0.003202 per run at the default inputs. $0.195 per 1M input tokens, $0.78 per 1M output tokens.

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| seed |  | — |  |
| stop |  | — |  |
| tools |  | — |  |
| top_p | 0 to 1 | — |  |
| reasoning |  | — | Reasoning effort; reasoning tokens count toward max_tokens. |
| max_tokens | 1 to 209715 | — | The most output tokens to generate. |

## What can you do with Mistral Small 4 in Perplexity?

Mistral Small 4 can turn supplied documents into explanations, comparisons, and multilingual replies. Give it a clear question, relevant evidence, and the response format you need.

### Trace an unfamiliar codebase

Follow a request from its entry point through validation, business logic, and storage. Ask for file references and a short explanation of each step.

### Investigate a failing test

Compare the failure output with the relevant implementation, identify a likely cause, and propose a focused fix. Use high reasoning when several components interact.

### Compare lengthy documents

Find conflicting requirements across policies or specifications. Request section references and separate explicit statements from interpretations.

### Turn support requests into records

Classify each request and extract its issue, requested action, and unresolved questions into a defined JSON structure.

### Draft multilingual support replies

Use supplied policy text to draft replies in English, French, German, or Spanish. Preserve product names and flag questions the policy does not answer.

### Explain charts

Analyze a supplied chart image, describe the visible trend, and identify the labels or values supporting the explanation. Ask it to mark unreadable details.

### Extract document details

Pull dates, organizations, obligations, and supporting passages from supplied documents into a consistent record for review.

### Compare constrained plans

Evaluate proposals against stated requirements, explain which constraints each satisfies, and list the missing information needed for a decision.

## Overview

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

## Mistral Small 4 in other agents

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

## More Perplexity guides

- [Mistral Small 3.2 24B](https://spiral.ai/integrations/perplexity/mistral-small-3-2-24b)
- [Mixtral 8x22B Instruct](https://spiral.ai/integrations/perplexity/mixtral-8x22b-instruct)
- [Mistral Small 3.1 24B Instruct](https://spiral.ai/integrations/perplexity/mistral-small-3-1-24b-instruct)
- [Moondream 3 Preview](https://spiral.ai/integrations/perplexity/moondream-3-preview)
- [Mistral Small 3](https://spiral.ai/integrations/perplexity/mistral-small-3)
- [Moondream 3.1](https://spiral.ai/integrations/perplexity/moondream-3-1)

## Questions about Mistral Small 4 in Perplexity

**Can Perplexity use Mistral Small 4?**

Yes. Add Spiral to Perplexity as an MCP server, and Perplexity can find and run Mistral Small 4 like any built-in tool.

**How much does Mistral Small 4 cost in Perplexity?**

Up to $0.003202 per run at the default inputs. $0.195 per 1M input tokens, $0.78 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Perplexity plan is separate.

**Do I need an OpenRouter account to use Mistral Small 4 in Perplexity?**

No. Spiral runs Mistral Small 4 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 4 in Perplexity?**

About two minutes: add a remote connector, then sign in to Spiral and approve.

**Does Perplexity wait while Mistral Small 4 runs?**

Mistral Small 4 runs in the background. Perplexity starts the run, checks on it and gives you the text, json when it's ready.

**What is Mistral Small 4 useful for?**

Use it for code exploration, document analysis, multilingual conversations, and problems that need several reasoning steps. It also supports structured extraction when you need a consistent result.

**When should I use deeper reasoning?**

Use high reasoning for difficult debugging, math, or comparisons with several constraints. For routine summaries and straightforward questions, start with reasoning turned off; high reasoning generates additional thinking tokens.

**How much text can Mistral Small 4 handle?**

Its context window is 256k tokens. Input and output share that allowance, so leave room for the answer instead of filling the entire window with source material.

**Which languages does Mistral Small 4 support?**

Mistral lists English, French, Spanish, German, Italian, Portuguese, Dutch, Chinese, Japanese, Korean, and Arabic. Specify the response language and provide any terminology you want preserved.

**Which temperature should I start with?**

Mistral recommends 0.7 with high reasoning. With reasoning turned off, its recommended range is 0.0 to 0.7, depending on the task.

