How to Use Pareto 26.10 Preview in Devin
Pareto 26.10 Preview is for testing research, coding, and agent workflows against the October preview. Unlike the stable Pareto 26.10 release, its launch notes explicitly allowed continued changes during preview.
Prompts to try with Pareto 26.10 Preview in Devin
Copy an example into Devin, or ask in your own words.
Use Pareto 26.10 Preview to review the current diff for bugs. For each finding, include the file, the triggering input, and the expected versus actual behavior.
Use Pareto 26.10 Preview to build a support-ticket extraction prompt from the examples in this project. Define the categories and add cases with missing order numbers.
Ask Pareto 26.10 Preview through Spiral to review the diff on this branch and list the three riskiest changes. Save the result in ./assets.
Get a second opinion from Pareto 26.10 Preview on this function, then tell me what the call cost.
Summarize these release notes with Pareto 26.10 Preview in five bullet points.
How to set up Pareto 26.10 Preview in Devin
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Add a custom MCP
Open Customize › MCPs. Click + and choose New MCP. Fill in:
- 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.
- Check it
On Spiral's configuration page, click Test tools. Devin lists Spiral's tools.
Use the Spiral skill
- Send Devin 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 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 Pareto 26.10 Preview
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": "unbiased/pareto-26.10-preview", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Devin finds Pareto 26.10 Preview on Spiral, shows the price and runs it.
Pareto 26.10 Preview providers and prices
Spiral runs Pareto 26.10 Preview through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.017077 per run at the default inputs. $1.04 per 1M input tokens, $4.16 per 1M output tokens.
Settings you can ask for
Ask Devin for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Top p | 0 to 1 | ||
| Max tokens | 1 to 131072 | The most output tokens to generate. On a reasoning model the reasoning tokens count toward it: a cap too low for the effort can end with no answer (finish_reason length). | |
| Temperature | 0 to 2 |
Every API field of Pareto 26.10 Preview is on its tool page: open its card above.
What can you do with Pareto 26.10 Preview in Devin?
Turn code, requirements, and test output into proposed changes and review findings with Pareto 26.10. Give the agent a bounded task and require evidence from the project before accepting a patch.
Research briefs
Turn supplied research into a brief with source references, disagreements, and unanswered questions.
Pull request review
Review a diff for concrete bugs and explain each finding with a reproducible example.
Support ticket extraction
Turn support messages into records containing the customer, order number, issue category, and missing details.
Inbox triage
Classify incoming messages by urgency and draft replies for review.
Proposal comparison
Compare supplied proposals against explicit requirements and identify assumptions that affect the decision.
Draft revision
Rewrite rough notes into concise messages while preserving names, dates, requests, and other supplied details.
More on Pareto 26.10 Preview and Devin
Overview
Pareto 26.10 Preview in other agents
More Devin guides
Questions about Pareto 26.10 Preview in Devin
Can Devin use Pareto 26.10 Preview?
Yes. Add Spiral to Devin as an MCP server, and Devin can find and run Pareto 26.10 Preview like any built-in tool.
How much does Pareto 26.10 Preview cost in Devin?
Up to $0.017077 per run at the default inputs. $1.04 per 1M input tokens, $4.16 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 Pareto 26.10 Preview in Devin?
No. Spiral runs Pareto 26.10 Preview on its own OpenRouter account. You sign in to Spiral once and never paste a key.
How long does it take to set up Pareto 26.10 Preview in Devin?
About two minutes: add a custom MCP, then sign in to Spiral and approve, then check it.
Does Devin wait while Pareto 26.10 Preview runs?
Pareto 26.10 Preview runs in the background. Devin starts the run, checks on it and gives you the text, json when it's ready.
What does Pareto mean by a blended model?
Pareto selects frontier and open models for each request, then blends their outputs into one answer. The response you receive combines work from several models.
How much can Pareto 26.10 read and write?
Pareto 26.10 and Pareto 26.10 Preview each list a 1,048,576-token context window and up to 131,072 completion tokens. These are token limits, not guaranteed word counts.
How does Pareto 26.10 differ from Pareto 26.10 Preview?
Pareto 26.10 is the stable release that followed the October preview. The preview launch notes allowed continued changes, while the listed context and completion limits are the same.
How much confidence should I put in Pareto's benchmark scores?
The published October results are preview measurements, not a new evaluation of the stable release. Use them to choose tasks to test, then check performance on your own examples.
How should I prompt Pareto for document work?
State the question, identify the source material, and specify the answer format and length. For extraction, define each label and tell it to leave missing information blank rather than infer it.
Start using Pareto 26.10 Preview in Devin
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Devin:
set up https://spiral.ai/SKILL.md


