Skip to main content
Muse Wonder GuideOfficial docs
Home / Getting started

Choose a clear first step with Muse Wonder

Use the official documentation for account setup and executable instructions. This guide helps you decide which access path to open first and what to check before you spend time or money.

Checked October 13, 2026

1. Choose an access path

Pick the path based on where you need the model to work. Choose the API path for a product or automation that makes programmatic requests. Choose the Muse Code path if you want Muse Wonder Full in the supported coding experience and can use an eligible plan.

Use the model pickerCompare the two paths

Make a first Wonder API request

Create an API key in the Meta Model API dashboard and enable billing if your account is eligible. Set MODEL_API_KEY in your local environment. For an OpenAI-compatible client, the base URL is https://api.meta.ai/v1 and Wonder's model ID is muse-wonder-1.0.

curl https://api.meta.ai/v1/responses \
  -H "Authorization: Bearer $MODEL_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"muse-wonder-1.0","input":"Explain this model in one sentence."}'

This example adapts the official quickstart request format to the documented Wonder model ID. It has not been executed by this site. Inspect the response's usage fields and actual billed amount before increasing your workload.

Install and start Muse Code

On macOS or Linux, Meta documents the installer below. On Windows, follow the PowerShell instructions in the official Muse Code guide.

curl -fsSL https://dev.meta.ai/install.sh | sh
muse --version
cd /path/to/your/project
muse

Complete browser sign-in during onboarding. For Full, confirm Power or Maximum eligibility, then select Full in the available model settings. Installing the CLI alone does not grant Full access.

Check the current setup details

Open the official quickstart for the current onboarding flow, credentials, request format, supported tools, and examples. If you are using Muse Code, open the official Muse Code documentation for its supported setup and plan requirements.

Read the official quickstartRead Muse Code docs

Account eligibility, available models and limits can change. Check the official setup and subscription pages before committing to a plan.

Evaluate one representative task

Start with our free coding workflow trial: download a small task with acceptance tests, then record your own time, retries and observed usage. For setup problems, follow the troubleshooting guide.

  1. Use a small, representative task to learn the interaction and output shape.
  2. For API work, set a token budget and review the input, cached-input, and output breakdown.
  3. For subscription work, check your current plan, quota, and regional availability in the official account experience.
  4. Keep private credentials and sensitive source material out of places where they do not belong.
  5. Review output before you rely on it in production, policy, or high-impact decisions.

Context capacity and subscription eligibility are product facts, not performance guarantees. Test the task you care about using the access path available to you.

Checked October 13, 2026