How to Use Gemini 3.1 Pro Preview in Muse Code
Gemini 3.1 Pro Preview suits complex coding, document synthesis, and tasks that require several reasoning steps. Its 1,048,576-token input limit makes it useful for examining large bodies of related material together.
Prompts to try with Gemini 3.1 Pro Preview in Muse Code
Copy an example into Muse Code, or ask in your own words.
Use Gemini 3.1 Pro Preview to trace the failing test through the relevant project files. Explain the cause, make the smallest supported fix, and run the focused tests.
Use Gemini 3.1 Pro Preview to compare this project's API client with the migration guide in the repository. List breaking changes, update the affected calls, and test the changed behavior.
Use Gemini 3.1 Pro Preview to review the queue implementation for duplicate processing and race conditions. Cite the relevant functions and add a regression test for each confirmed defect.
Use Gemini 3.1 Pro Preview to compare two algorithms for the bottleneck identified by the project's benchmark. Explain time and memory complexity, implement the better candidate, and rerun the benchmark.
Use Gemini 3.1 Pro Preview to create an animated SVG explaining this project's request lifecycle. Follow the existing visual style, label each stage, and preview the result.
How to set up Gemini 3.1 Pro Preview 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 Gemini 3.1 Pro 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": "vertex", "model": "google/gemini-3.1-pro-preview", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Muse Code finds Gemini 3.1 Pro Preview on Spiral, shows the price and runs it.
Gemini 3.1 Pro Preview providers and prices
Spiral runs Gemini 3.1 Pro Preview through Vertex AI. Open a card for today's price, with every fee included.
Up to $0.063989 per run at the default inputs. $2.60 per 1M input tokens, $0.26 per 1M cached input tokens, $0.4875 per 1M cache write tokens, $15.60 per 1M output tokens; for a prompt above 200,000 tokens, $5.20 per 1M input tokens, $0.52 per 1M cached input tokens, $0.975 per 1M cache write tokens, $23.40 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 | ||
| Max tokens | 1 to 65536 | The most output tokens to generate. | |
| Temperature | 0 to 2 | ||
| Presence penalty | a number from -2 | ||
| Reasoning effort | low, medium, or high | high | How much the model reasons. |
| Frequency penalty | a number from -2 |
Every API field of Gemini 3.1 Pro Preview is on its tool page: open its card above.
What can you do with Gemini 3.1 Pro Preview in Muse Code?
Gemini 3.1 Pro can help trace behavior across related code and turn a complex change into a concrete implementation. Use the coding agent to supply relevant files, then check the proposed changes against tests and project requirements.
Trace bugs across a codebase
Use Gemini 3.1 Pro Preview to follow a failing behavior through related modules, propose a focused fix, and identify tests that would confirm it.
Plan a software migration
Compare existing code with supplied migration documentation. Ask for dependencies, likely breaking changes, and a staged implementation plan.
Compare lengthy specifications
Review several specifications together to identify conflicting requirements, missing definitions, and decisions that need an owner.
Build an evidence-based research brief
Synthesize supplied source text into findings, disagreements, and unanswered questions. Require a supporting passage for each finding.
Extract records from unstructured text
Turn support messages or operational notes into structured records. Define how missing values should appear and validate the extracted values against the source.
Develop and compare algorithms
Describe the constraints and representative inputs, then ask for candidate algorithms, complexity analysis, and edge cases to test.
Create animated SVG explanations
Generate an animated SVG that explains a process, such as a queue draining or a request passing through a cache. Specify labels, timing, and the intended audience.
Summarize recurring operational changes
Compare supplied current and previous reports to produce a brief of new issues, resolved issues, and contradictions that need review.
More on Gemini 3.1 Pro Preview and Muse Code
Gemini 3.1 Pro Preview in other agents
Questions about Gemini 3.1 Pro Preview in Muse Code
Can Muse Code use Gemini 3.1 Pro Preview?
Yes. Add Spiral to Muse Code as an MCP server, and Muse Code can find and run Gemini 3.1 Pro Preview like any built-in tool.
How much does Gemini 3.1 Pro Preview cost in Muse Code?
Up to $0.063989 per run at the default inputs. $2.60 per 1M input tokens, $0.26 per 1M cached input tokens, $0.4875 per 1M cache write tokens, $15.60 per 1M output tokens; for a prompt above 200,000 tokens, $5.20 per 1M input tokens, $0.52 per 1M cached input tokens, $0.975 per 1M cache write tokens, $23.40 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 a Vertex AI account to use Gemini 3.1 Pro Preview in Muse Code?
No. Spiral runs Gemini 3.1 Pro Preview on its own Vertex AI account. You sign in to Spiral once and never paste a key.
How long does it take to set up Gemini 3.1 Pro Preview in Muse Code?
About two minutes: install Muse Code, then ask Muse to add the server, then sign in.
Does Muse Code wait while Gemini 3.1 Pro Preview runs?
Gemini 3.1 Pro Preview runs in the background. Muse Code starts the run, checks on it and gives you the text, json when it's ready.
What is Gemini 3.1 Pro best suited for?
Use it for complex coding, algorithm development, and questions that require comparing substantial supporting material. Give it the relevant evidence and define the result you need.
How much can Gemini 3.1 Pro read and write?
Gemini 3.1 Pro Preview accepts up to 1,048,576 input tokens and generates up to 65,536 output tokens. Treat these as token limits, not fixed page or word counts.
Will Gemini 3.1 Pro find every detail in a long document?
A large context window does not guarantee complete recall. Google reports lower retrieval accuracy when a request asks for multiple specific details, so ask focused questions and verify the cited passages.
How should I structure a long prompt?
Put the supporting material first and your question at the end. Ask for a concrete deliverable, such as a comparison table with evidence for each conclusion.
Should I lower the temperature for analytical work?
Start with the default sampling settings. Google warns that changing these defaults on earlier Gemini 3 models can degrade reasoning performance.
Start using Gemini 3.1 Pro Preview 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


