How to Use Mistral NeMo in Instinct
Mistral NeMo is a 12B text model for multilingual conversations, long-document summaries, classification, and code generation. It supports up to 128k tokens of context, and Mistral recommends a temperature of 0.3.
Prompts to try with Mistral NeMo in Instinct
Copy an example into Instinct, or ask in your own words.
Use Mistral NeMo with a temperature of 0.3 on each daily run to summarize the supplied support messages. Report recurring issues, include message IDs as evidence, and separate observations from suggested actions.
Use Mistral NeMo on each run to label incoming requests as billing, delivery, technical problem, or other. Include the request ID and a brief reason, and flag ambiguous requests for review.
Use Mistral NeMo each week to turn the supplied meeting transcripts into a digest of decisions, assigned actions, and unresolved questions. Preserve stated owners and dates; mark missing details as unstated.
Use Mistral NeMo on each run to draft French and German replies to the supplied customer questions using only the approved policy text. Flag questions the policy does not answer for human review.
Use Mistral NeMo on each scheduled run to summarize the supplied release notes in plain English. List changes, affected features, and any explicit migration instructions without adding unsupported details.
How to set up Mistral NeMo 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 NeMo
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-nemo", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Instinct finds Mistral NeMo on Spiral, shows the price and runs it.
Mistral NeMo providers and prices
Spiral runs Mistral NeMo through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.000162 per run at the default inputs. $0.0377 per 1M input tokens, $0.039 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 | ||
| Min p | 0 to 1 | ||
| Top k | a whole number from 0 | ||
| Top p | 0 to 1 | ||
| Logprobs | on or off | ||
| Max tokens | 1 to 16384 | The most output tokens to generate. | |
| Temperature | 0 to 2 | ||
| Top logprobs | 0 to 20 | ||
| Presence penalty | -2 to 2 | ||
| Frequency penalty | -2 to 2 | ||
| Repetition penalty | 0 to 2 |
Every API field of Mistral: Mistral Nemo is on its tool page: open its card above.
What can you do with Mistral NeMo in Instinct?
Produce recurring digests, request labels, and multilingual reply drafts with Mistral NeMo. Give each run the current source text and a fixed answer format so the results are easy to review.
Long-report summaries
Ask Mistral NeMo to turn supplied report text into findings, decisions, and open questions. Request a supporting passage for each finding.
Multilingual support drafts
Use Mistral NeMo to draft replies in French, German, Spanish, or Japanese from a supplied customer question and approved policy text.
Support request classification
Give Mistral NeMo a fixed set of categories and examples, then ask it to label each request and flag cases that need review.
Code and test drafts
Ask Mistral NeMo to draft a function and tests from a written specification, with expected behavior and edge cases stated explicitly.
Product text translation
Have Mistral NeMo translate product descriptions or help text while preserving product names, measurements, and a supplied terminology list.
Meeting digests
Ask Mistral NeMo to summarize a supplied transcript into decisions, assigned actions, and unresolved questions without inventing owners or dates.
Follow-up explanations
Use Mistral NeMo to explain supplied technical material, then ask follow-up questions about specific passages or examples.
More on Mistral NeMo and Instinct
Overview
Mistral NeMo in other agents
Questions about Mistral NeMo in Instinct
Can Instinct use Mistral NeMo?
Yes. Add Spiral to Instinct as an MCP server, and Instinct can find and run Mistral NeMo like any built-in tool.
How much does Mistral NeMo cost in Instinct?
Up to $0.000162 per run at the default inputs. $0.0377 per 1M input tokens, $0.039 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 NeMo in Instinct?
You need no OpenRouter account. Instinct uses one Spiral API key for every tool, and Spiral runs Mistral NeMo on its own OpenRouter account.
How long does it take to set up Mistral NeMo in Instinct?
About two minutes: create an API key, then send Instinct this message, then give Instinct the key.
Does Instinct wait while Mistral NeMo runs?
Mistral NeMo runs in the background. Instinct starts the run, checks on it and gives you the text, json when it's ready.
What is Mistral NeMo good at?
Mistral NeMo supports long-document summarization, classification, translation, code generation, and conversations with follow-up questions. Useful starting tasks include summarizing a report, labeling support requests, and drafting code from a written specification.
How much text can Mistral NeMo handle?
Mistral documents a context window of up to 128k tokens. Input and output share that budget, so leave room for the answer instead of filling it entirely with source text.
Which languages work well with Mistral NeMo?
Mistral highlights English, French, German, Spanish, Italian, Portuguese, Chinese, Japanese, Korean, Arabic, and Hindi. For translation, specify the target language, audience, and terms that must stay unchanged.
What temperature should I use?
Start at 0.3, Mistral’s explicit recommendation for NeMo. Use that as your baseline when comparing prompts for summaries, classification, or code.
How should I structure a Mistral NeMo prompt?
Separate the task, source text, constraints, and desired answer format into clear sections. Add a worked example when you need consistent labels or a particular response structure.
Start using Mistral NeMo 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


