How to Use Muse Spark 1.3 in Amp
Muse Spark 1.3 suits coding tasks and research that require several steps, tool calls, and a long working context. Compared with Muse Spark 1.2, Meta reports fewer unnecessary turns, less verbose responses, and cleaner code.
Prompts to try with Muse Spark 1.3 in Amp
Copy an example into Amp, or ask in your own words.
Use Muse Spark 1.3 to trace the failing test in this repository through the relevant modules. Explain the cause, make the smallest justified fix, and run the focused regression check.
Use Muse Spark 1.3 with high reasoning effort to plan this repository's dependency upgrade. Identify breaking changes, order the edits, and define a verification check for each stage before changing code.
Use Muse Spark 1.3 to compare the implementation with the requirements in this project's documentation. List each mismatch with its file location and propose a focused correction.
Use Muse Spark 1.3 to inspect the project's configuration loader and propose a JSON schema for its accepted configuration. Identify where the code and documentation disagree.
Use Muse Spark 1.3 to review the current diff against the earlier task requirements. Check each requirement separately, report the evidence, and identify anything still unverified.
How to set up Muse Spark 1.3 in Amp
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Add the server
No header and no client ID or secret: Amp registers itself with Spiral.
amp mcp add spiral https://api.spiral.ai/mcp
- Sign in
Start amp. Amp opens Spiral's sign-in page in your browser: sign in, pick the organization and click Authorize. No API key.
- Check it
amp mcp doctor
Use the Spiral skill
- Send Amp 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 Muse Spark 1.3
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": "meta/muse-spark-1.3", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Amp finds Muse Spark 1.3 on Spiral, shows the price and runs it.
Muse Spark 1.3 providers and prices
Spiral runs Muse Spark 1.3 through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.022688 per run at the default inputs. $1.625 per 1M input tokens, $5.525 per 1M output tokens.
Settings you can ask for
Ask Amp for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Top k | a whole number from 0 | ||
| Top p | 0 to 1 | ||
| Max tokens | 1 to 943718 | The most output tokens to generate. | |
| Temperature | 0 to 2 | ||
| Reasoning effort | minimal, low, medium, high, xhigh, or max | medium | How much the model reasons. |
| Repetition penalty | 0 to 2 |
Every API field of Meta: Muse Spark 1.3 is on its tool page: open its card above.
What can you do with Muse Spark 1.3 in Amp?
Muse Spark 1.3 can help carry a coding task through investigation, edits, and verification across several steps. Give it the repository context, acceptance criteria, and a specific check for completion.
Trace a bug across a codebase
Follow a failing request through related modules, identify the cause, and propose a focused patch with a regression check.
Plan a software migration
Map dependencies, sequence changes, and define acceptance checks for each stage of a framework or library upgrade.
Compare conflicting research
Turn several sources into a comparison that distinguishes agreements, contradictions, and questions the evidence leaves open.
Reconcile a changing brief
Compare original requirements with later feedback and draft a revised plan that makes unresolved decisions explicit.
Extract records into JSON
Pull names, dates, locations, and product details from supplied text into a defined structure for downstream processing.
Explain a chart
Ask questions about a supplied chart and request a written explanation tied to its visible labels and values.
Prepare a recurring change brief
Compare current source text with an earlier summary and draft a concise account of what changed, what conflicts, and what needs attention.
More on Muse Spark 1.3 and Amp
Overview
Muse Spark 1.3 in other agents
Questions about Muse Spark 1.3 in Amp
Can Amp use Muse Spark 1.3?
Yes. Add Spiral to Amp as an MCP server, and Amp can find and run Muse Spark 1.3 like any built-in tool.
How much does Muse Spark 1.3 cost in Amp?
Up to $0.022688 per run at the default inputs. $1.625 per 1M input tokens, $5.525 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Amp plan is separate.
Do I need an OpenRouter account to use Muse Spark 1.3 in Amp?
No. Spiral runs Muse Spark 1.3 on its own OpenRouter account. You sign in to Spiral once and never paste a key.
How long does it take to set up Muse Spark 1.3 in Amp?
About two minutes: add the server, then sign in, then check it.
Does Amp wait while Muse Spark 1.3 runs?
Muse Spark 1.3 runs in the background. Amp starts the run, checks on it and gives you the text, json when it's ready.
What is Muse Spark 1.3 good at?
It suits coding and research tasks that need several steps, repeated tool use, and reasoning across substantial context. Good starting points include debugging, migration planning, and comparing evidence from several sources.
How does Muse Spark 1.3 differ from Muse Spark 1.2?
Meta reports fewer unnecessary turns, less verbose responses, and a cleaner coding style. Test those improvements on a representative task with explicit acceptance criteria.
How much context can Muse Spark 1.3 handle?
Meta lists a 1,048,576-token context window. It returns text; organize lengthy requests around named sources, clear questions, and a defined final deliverable.
When should I increase reasoning effort?
Use higher effort for complex code generation, architecture choices, and planning dependent steps. Leave enough output room for both reasoning and the answer, since reasoning consumes part of that allowance.
What temperature should I use?
Meta recommends 1.0 for most Muse Spark work. Start there and make the requested audience, structure, and constraints explicit before experimenting.
Start using Muse Spark 1.3 in Amp
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Amp:
set up https://spiral.ai/SKILL.md


