How to Use Muse Spark 1.3 in Instinct
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 Instinct
Copy an example into Instinct, or ask in your own words.
Use Muse Spark 1.3 on each run to compare the current product announcements supplied to you with the previous run's summary. Report changed capabilities, conflicting statements, and the supporting source for each change.
Use Muse Spark 1.3 each morning to reconcile the project updates supplied for this run. Produce a brief covering changed deadlines, dependencies, and conflicting instructions.
Use Muse Spark 1.3 on each scheduled run to extract names, dates, locations, and product details from the supplied records. Return JSON, preserve record identifiers, and use null for absent values.
Use Muse Spark 1.3 each week to compare the supplied research summaries. Group agreements and contradictions, cite the relevant source text, and draft five questions for further investigation.
Use Muse Spark 1.3 with high reasoning effort each week to review the supplied plan and progress updates. Reorder dependent tasks where justified and explain which evidence supports each proposed change.
How to set up Muse Spark 1.3 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 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: Instinct 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 Instinct 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 Instinct?
Muse Spark 1.3 can prepare recurring comparisons, reconcile incoming updates, and turn supplied records into structured summaries. Define what each run should compare and what evidence belongs in the result.
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 Instinct
Overview
Muse Spark 1.3 in other agents
Questions about Muse Spark 1.3 in Instinct
Can Instinct use Muse Spark 1.3?
Yes. Add Spiral to Instinct as an MCP server, and Instinct can find and run Muse Spark 1.3 like any built-in tool.
How much does Muse Spark 1.3 cost in Instinct?
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 Instinct plan is separate.
Do I need an OpenRouter account to use Muse Spark 1.3 in Instinct?
You need no OpenRouter account. Instinct uses one Spiral API key for every tool, and Spiral runs Muse Spark 1.3 on its own OpenRouter account.
How long does it take to set up Muse Spark 1.3 in Instinct?
About two minutes: create an API key, then send Instinct this message, then give Instinct the key.
Does Instinct wait while Muse Spark 1.3 runs?
Muse Spark 1.3 runs in the background. Instinct 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 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


