[Integrations](https://spiral.ai/integrations) › [Instinct](https://spiral.ai/integrations/instinct) › [Mistral Small 3](https://spiral.ai/integrations/instinct/mistral-small-3)

# How to Use Mistral Small 3 in Instinct

Use Mistral Small 3 for multilingual support replies, text classification, and JSON extraction. Its 32k context is smaller than Mistral Small 3.1's 128k window.

|  |  |
| --- | --- |
| Model | Mistral Small 3 |
| Available through | OpenRouter |
| Category | LLM |
| Price | Up to $0.000429 per run at the default inputs. $0.065 per 1M input tokens, $0.104 per 1M output tokens. |
| Returns | text, json |
| Runs | In the background |
| Setup | About 2 minutes, with a Spiral API key |

## Prompts to try with Mistral Small 3 in Instinct

- Every weekday, use Mistral Small 3 to summarize the customer feedback collected since the previous run. Return the five main themes with supporting quotes and the original feedback identifiers.
- Each morning, use Mistral Small 3 to classify newly collected support messages as delivery, returns, product question, or unclear. Use low randomness and include one reason for each label.
- Every Friday, use Mistral Small 3 to turn the week's meeting notes into decisions, unfinished actions, and open questions. Preserve explicitly stated owners and deadlines; mark missing details as unknown.
- Each day, use Mistral Small 3 to draft replies to new French and Spanish customer messages using the supplied support policy. Keep replies in the customer's language and flag missing policy information.
- Every morning, use Mistral Small 3 to compare the latest collected release notes with the previous collected version. Summarize added, changed, and removed behavior in under 300 words, citing the supplied passages.

## How to set up Mistral Small 3 in Instinct

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

### Use the Spiral skill

1. **Send Instinct 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 Spiral's MCP server

1. **Create an API key**: In the console, open API keys and create a key for Instinct. Copy it.

2. **Send Instinct this message**: Instinct reads Spiral's skill and asks for your API key.

```text
set up https://spiral.ai/SKILL.md
```

3. **Give Instinct the key**: When Instinct asks for the key, add it where Instinct keeps secrets, never in the chat itself.

### 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**: 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-24b-instruct-2501",
    "input": {
      "messages": [
        {
          "role": "system"
        }
      ]
    }
  }'
```

Then paste one of the prompts above: Instinct finds Mistral Small 3 on Spiral, shows the price and runs it.

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

## Mistral Small 3 tools and prices

- [Mistral: Mistral Small 3](https://spiral.ai/tools/openrouter/mistralai/mistral-small-24b-instruct-2501) (OpenRouter): Up to $0.000429 per run at the default inputs. $0.065 per 1M input tokens, $0.104 per 1M output tokens.

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| seed |  | — |  |
| stop |  | — |  |
| min_p | 0 to 1 | — |  |
| top_k | 0 to … | — |  |
| top_p | 0 to 1 | — |  |
| max_tokens | 1 to 16384 | — | The most output tokens to generate. |

## What can you do with Mistral Small 3 in Instinct?

Produce recurring digests, feedback labels, and support drafts from text collected by a scheduled agent. Keep each Mistral Small 3 task focused on the current batch and a fixed response format.

### Support reply drafts

Draft an answer from a customer's message and the relevant policy excerpt. Require the draft to identify questions the policy does not answer.

### Feedback classification

Label customer feedback by topic and sentiment. Define the allowed labels and request one supporting phrase for each decision.

### Structured text extraction

Turn order messages or intake notes into JSON records. Specify the required keys and use null when the source omits a value.

### Multilingual customer communication

Translate support replies between supported languages while preserving product names, identifiers, and the meaning of policy terms.

### Document and meeting digests

Summarize selected document text or meeting notes into decisions, action items, and open questions. Keep the supplied material within the 32k context window.

### Focused code assistance

Explain a function, suggest a small patch, or draft tests from relevant code excerpts. Check generated changes with the project's existing tools.

## Overview

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

## Mistral Small 3 in other agents

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

## More Instinct guides

- [Mistral Saba](https://spiral.ai/integrations/instinct/mistral-saba)
- [Mistral Small 3.1 24B Instruct](https://spiral.ai/integrations/instinct/mistral-small-3-1-24b-instruct)
- [Mistral NeMo](https://spiral.ai/integrations/instinct/mistral-nemo)
- [Mistral Small 3.2 24B](https://spiral.ai/integrations/instinct/mistral-small-3-2-24b)
- [Mistral Medium 3.5](https://spiral.ai/integrations/instinct/mistral-medium-3-5)
- [Mistral Small 4](https://spiral.ai/integrations/instinct/mistral-small-4)

## Questions about Mistral Small 3 in Instinct

**Can Instinct use Mistral Small 3?**

Yes. Add Spiral to Instinct as an MCP server, and Instinct can find and run Mistral Small 3 like any built-in tool.

**How much does Mistral Small 3 cost in Instinct?**

Up to $0.000429 per run at the default inputs. $0.065 per 1M input tokens, $0.104 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Instinct plan is separate.

**Do I need an OpenRouter account to use Mistral Small 3 in Instinct?**

You need no OpenRouter account. Instinct uses one Spiral API key for every tool, and Spiral runs Mistral Small 3 on its own OpenRouter account.

**How long does it take to set up Mistral Small 3 in Instinct?**

About two minutes: create an API key, then send Instinct this message, then give Instinct the key.

**Does Instinct wait while Mistral Small 3 runs?**

Mistral Small 3 runs in the background. Instinct starts the run, checks on it and gives you the text, json when it's ready.

**What is Mistral Small 3 good for?**

Start with support replies, text classification, multilingual assistance, and structured extraction. Mistral specifically identifies customer service and sentiment and feedback analysis as practical uses.

**How much context can Mistral Small 3 handle?**

Mistral Small 3 has a 32k context window. Keep document selections and conversation history focused; the 128k window belongs to Mistral Small 3.1.

**Which languages does Mistral Small 3 support?**

The model card lists dozens of supported languages, including English, French, German, Spanish, Italian, and Chinese. State the target language and intended audience when requesting a translation or reply.

**Can Mistral Small 3 return JSON?**

Yes, its model card documents JSON output and native function calling. For extraction, specify the exact keys, allowed labels, and how to represent missing information.

**What temperature should I start with?**

Start around 0.15, the value in Mistral Small 3's published generation configuration. Pair it with a narrow task and a clear response format.

