Overview

• Download and extract extension
  ⬇ Download muse2api-extension.zip View Included Files
• Load unpacked in browser
  Open chrome://extensions (or Edge / Brave equivalent) → Enable Developer mode → Click Load unpacked → Select the extracted folder.
• Log in to muse.ai and sync
  Log into https://muse.ai/ in your browser until the chat screen appears. Click the extension icon in your toolbar, input these two values, and click "Read and Import":
  BASE URL

  Copy

  API Key

  Copy

• Log into https://muse.ai/ and stay on the chat view.
• Press F12 to open Developer Tools, then switch to the Network tab.
• Press F5 to reload the page to record HTTP requests.
• Select any request to muse.ai. In the right pane, locate Request Headers.
• Find the header line starting with cookie:. Copy the complete value after the colon.
• Paste into Option 3 below and click Import.

Copycurl -X POST "BASE_URL/v1/chat/completions" \
  -H "Authorization: Bearer $MUSE2API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "muse-spark",
    "messages": [{"role": "user", "content": "Explain quicksort in Python"}],
    "stream": false
  }'

Copycurl -X POST "BASE_URL/v1/images/generations" \
  -H "Authorization: Bearer $MUSE2API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "muse-image",
    "prompt": "Futuristic skyline with flying vehicles at sunset",
    "size": "16:9",
    "response_format": "url"
  }'

Copy# 1. Initiate video generation task
curl -X POST "BASE_URL/v1/videos" \
  -H "Authorization: Bearer $MUSE2API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "A serene waterfall in the jungle", "duration": 5, "size": "16:9"}'

# 2. Poll task status until status is succeeded
curl "BASE_URL/v1/videos/task_id" -H "Authorization: Bearer $MUSE2API_KEY"

Copyfrom openai import OpenAI

client = OpenAI(base_url="http://YOUR_SERVER:18610/v1", api_key="m2a_YOUR_KEY")

# Chat completion
response = client.chat.completions.create(
    model="muse-spark",
    messages=[{"role": "user", "content": "Hello world!"}]
)
print(response.choices[0].message.content)

# Image generation
image = client.images.generate(
    model="muse-image",
    prompt="Cyberpunk neon city street"
)
print(image.data[0].url)