How to Use MiMo-V2.5-Pro in Muse Code
MiMo-V2.5-Pro targets complex software engineering and tasks requiring many tool calls. Choose it over MiMo-V2.5 when a project needs sustained execution across a longer sequence of steps.
Prompts to try with MiMo-V2.5-Pro in Muse Code
Copy an example into Muse Code, or ask in your own words.
Use MiMo-V2.5-Pro to plan the database migration described in this project's requirements. Inspect the schema and callers, map dependencies, and propose staged changes with verification and rollback steps. Present the plan before editing.
Use MiMo-V2.5-Pro to implement pagination across this project's API and client. First identify existing conventions and acceptance criteria, then work component by component. Test empty results, page boundaries, and invalid inputs.
Use MiMo-V2.5-Pro to trace how authentication state moves through this codebase. Produce a dependency map, identify inconsistent expiration handling, and propose a focused correction with tests for expired and missing sessions.
Ask MiMo-V2.5-Pro 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 MiMo-V2.5-Pro on this function, then tell me what the call cost.
How to set up MiMo-V2.5-Pro in Muse Code
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Install Muse Code
curl -fsSL https://dev.meta.ai/install.sh | bash
- Ask Muse to add the server
Start muse and ask:
Add the Spiral MCP server at https://api.spiral.ai/mcp
- Sign in
In your terminal, run:
muse mcp login spiral
Use the Spiral skill
- Send Muse Code 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 MiMo-V2.5-Pro
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": "xiaomi/mimo-v2.5-pro", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Muse Code finds MiMo-V2.5-Pro on Spiral, shows the price and runs it.
MiMo-V2.5-Pro providers and prices
Spiral runs MiMo-V2.5-Pro through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.009607 per run at the default inputs. $0.624 per 1M input tokens, $2.34 per 1M output tokens.
Settings you can ask for
Ask Muse Code for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Seed | a whole number | ||
| Top p | 0 to 1 | ||
| Logprobs | on or off | ||
| Max tokens | 1 to 235929 | The most output tokens to generate. | |
| Temperature | 0 to 2 | ||
| Top logprobs | 0 to 20 |
Every API field of Xiaomi: MiMo-V2.5-Pro is on its tool page: open its card above.
What can you do with MiMo-V2.5-Pro in Muse Code?
Turn a bug report or feature specification into code changes and verification with MiMo-V2.5. Give the agent acceptance checks and ask it to report the evidence for completion.
Trace and fix a software bug
Work from a reproducible failure, trace the relevant code, propose a focused fix, and check it against a regression test.
Plan a code migration
Turn the current architecture and target requirements into an ordered migration plan with dependencies, verification steps, and rollback points.
Compare long specifications
Compare two text specifications, identify changed requirements, and attach section references to each difference.
Build an evidence-based research brief
Combine supplied source text into a brief that distinguishes supported findings, conflicting evidence, and unanswered questions.
Extract action items from meeting text
Convert a transcript into structured tasks with owners, deadlines, and supporting excerpts. Leave unspecified details empty.
Triage support conversations
Classify supplied conversation text using fixed categories, extract the reported problem, and draft a response grounded in the supplied policy.
Review a sequence of operational events
Read timestamped event text, construct a timeline, and separate observed failures from possible causes and recommended checks.
Implement a feature across several components
Translate a feature specification into coordinated code changes, track remaining requirements, and verify behavior at each stage. Prefer MiMo-V2.5-Pro for extended work.
More on MiMo-V2.5-Pro and Muse Code
Overview
MiMo-V2.5-Pro in other agents
Questions about MiMo-V2.5-Pro in Muse Code
Can Muse Code use MiMo-V2.5-Pro?
Yes. Add Spiral to Muse Code as an MCP server, and Muse Code can find and run MiMo-V2.5-Pro like any built-in tool.
How much does MiMo-V2.5-Pro cost in Muse Code?
Up to $0.004653 per run at the default inputs. $0.5655 per 1M input tokens, $1.131 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Muse Code plan is separate.
Do I need an OpenRouter account to use MiMo-V2.5-Pro in Muse Code?
No. Spiral runs MiMo-V2.5-Pro on its own OpenRouter account. You sign in to Spiral once and never paste a key.
How long does it take to set up MiMo-V2.5-Pro in Muse Code?
About two minutes: install Muse Code, then ask Muse to add the server, then sign in.
Does Muse Code wait while MiMo-V2.5-Pro runs?
MiMo-V2.5-Pro runs in the background. Muse Code starts the run, checks on it and gives you the text, json when it's ready.
What is MiMo-V2.5 good at?
Use MiMo-V2.5 for document analysis, coding, and tasks that combine reasoning with tool use. MiMo-V2.5-Pro puts more emphasis on complex software projects and extended sequences of actions.
How much context can MiMo-V2.5 handle?
Xiaomi publishes a 1M-token context window for both MiMo-V2.5 and MiMo-V2.5-Pro. For long documents, ask for section references beside conclusions so you can check the evidence.
Can MiMo-V2.5 write a million-token answer?
Xiaomi lists a maximum output length of 128K tokens for both models, separate from the 1M-token context window. For a long deliverable, request an outline first and draft one section at a time.
How should I prompt MiMo-V2.5 for structured results?
Define the exact fields, data types, and nesting you want, then provide a completed example. State how to handle missing information, such as returning null instead of guessing.
Will lowering temperature make thinking mode more predictable?
Xiaomi documents that thinking mode uses temperature 1.0 and top-p 0.95 for both models, overriding custom values. Start by tightening the instructions, allowed answers, and examples.
Start using MiMo-V2.5-Pro in Muse Code
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Muse Code:
set up https://spiral.ai/SKILL.md


