How to Use Mistral Medium 3.1 in Instinct
Mistral Medium 3.1 suits coding, document questions, and structured extraction. It updates Mistral Medium 3 with improvements to tone and performance.
Prompts to try with Mistral Medium 3.1 in Instinct
Copy an example into Instinct, or ask in your own words.
Each morning, use Mistral Medium 3.1 to extract order details from the new order messages supplied to this run. Return one JSON record per message, preserve its source identifier, use null for missing details, and choose low randomness.
Every weekday, use Mistral Medium 3.1 to draft replies for the unanswered support requests supplied to this run. Use only the accompanying policy text, include supporting passages, and flag requests that need more information.
Each week, use Mistral Medium 3.1 to compare the previous and current product specification texts supplied to this run. Report changed requirements, conflicts, and questions for review, with short supporting excerpts.
Each morning, use Mistral Medium 3.1 to prepare French and German drafts of the approved customer announcements supplied to this run. Follow the supplied glossary, preserve dates and product names, and flag ambiguous wording.
Every Friday, use Mistral Medium 3.1 to examine the technical incident descriptions supplied to this run. For each incident, propose regression test cases with inputs and expected results, and identify missing details needed to make the tests executable.
How to set up Mistral Medium 3.1 in Instinct
Pick the way that suits you. Each takes about two minutes.
Use the Spiral skill
- Send Instinct 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 Spiral's MCP server
- Create an API key
In the console, open API keys and create a key for Instinct. Copy it.
- Send Instinct this message
Instinct reads Spiral's skill and asks for your API key.
set up https://spiral.ai/SKILL.md
- 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
- 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 Mistral Medium 3.1
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": "mistralai/mistral-medium-3.1", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Instinct finds Mistral Medium 3.1 on Spiral, shows the price and runs it.
Mistral Medium 3.1 providers and prices
Spiral runs Mistral Medium 3.1 through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.011736 per run at the default inputs. $0.572 per 1M input tokens, $2.86 per 1M output tokens.
Settings you can ask for
Ask Instinct for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Seed | a whole number | ||
| Top p | 0 to 1 | ||
| Max tokens | 1 to 104857 | The most output tokens to generate. | |
| Temperature | 0 to 2 | ||
| Presence penalty | -2 to 2 | ||
| Frequency penalty | -2 to 2 |
Every API field of Mistral: Mistral Medium 3.1 is on its tool page: open its card above.
What can you do with Mistral Medium 3.1 in Instinct?
Turn recurring batches of supplied text into reviewable records and technical comparisons with Mistral Medium 3.1. Define the output structure and evidence requirements once so each scheduled run produces results a person can check.
Review a code change
Give Mistral Medium 3.1 a diff and the intended behavior. Ask for concrete failure cases, a suggested correction, and tests that would expose the problem.
Turn a bug report into regression tests
Supply the failing behavior and relevant code. Request tests for the reported failure, boundary conditions, and the expected result after a fix.
Answer questions about technical documents
Provide the document text and ask targeted questions about requirements, exceptions, or dependencies. Request a supporting passage for each answer and an explicit acknowledgment of missing evidence.
Extract records from business text
Convert supplied order emails, service requests, or meeting notes into structured records. Define the required structure and use null for information the source does not contain.
Localize customer communications
Draft French, German, Spanish, or Italian versions of supplied customer messages. Provide a terminology guide and ask to preserve product names, numbers, and commitments.
Prepare support reply drafts
Combine a customer's message with the relevant policy text to draft a response. Ask for the supporting policy passage and flag questions the supplied information cannot answer.
Explain a technical calculation
Supply the problem, assumptions, and units. Request a proposed solution, intermediate calculations, and checks that a person or calculator can verify.
Compare requirements across documents
Provide two specification texts and ask for changed requirements, conflicts, and unanswered questions. Request short supporting excerpts so a reviewer can check each finding.
More on Mistral Medium 3.1 and Instinct
Overview
Mistral Medium 3.1 in other agents
Questions about Mistral Medium 3.1 in Instinct
Can Instinct use Mistral Medium 3.1?
Yes. Add Spiral to Instinct as an MCP server, and Instinct can find and run Mistral Medium 3.1 like any built-in tool.
How much does Mistral Medium 3.1 cost in Instinct?
Up to $0.010668 per run at the default inputs. $0.52 per 1M input tokens, $2.60 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 Medium 3.1 in Instinct?
You need no OpenRouter account. Instinct uses one Spiral API key for every tool, and Spiral runs Mistral Medium 3.1 on its own OpenRouter account.
How long does it take to set up Mistral Medium 3.1 in Instinct?
About two minutes: create an API key, then send Instinct this message, then give Instinct the key.
Does Instinct wait while Mistral Medium 3.1 runs?
Mistral Medium 3.1 runs in the background. Instinct starts the run, checks on it and gives you the text, json when it's ready.
What is Mistral Medium 3.1 good at?
Coding and STEM reasoning are documented strengths. Practical tasks include reviewing code, explaining technical problems, and extracting structured information from supplied text.
How much text can Mistral Medium 3.1 handle?
Its context window is 131,072 tokens, commonly labeled 128k. Input and generated output share that space, so leave room for the answer when supplying long documents.
Can I rely on Mistral Medium 3.1 to return the exact JSON structure I need?
JSON mode ensures valid JSON but does not guarantee your required structure. Specify the expected format, prefer schema-constrained output when available, and validate the result before using it.
Can Mistral Medium 3.1 work in languages besides English?
Yes. Mistral's language documentation lists French, German, Spanish, Portuguese, Italian, Dutch, Polish, and Czech, among others; test your terminology and tone with representative examples.
Which temperature should I use for Mistral Medium 3.1?
Start around 0.2 for extraction and focused technical answers. Try around 0.7 when you want more variation in drafts; these are starting points from Mistral's general guidance.
Start using Mistral Medium 3.1 in Instinct
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Instinct:
set up https://spiral.ai/SKILL.md


