Integration guide

1. Available models

2. Chat / code

Copy# Non-streaming
curl -X POST "BASE/v1/chat/completions" \
  -H "Authorization: Bearer $MUSE2API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"muse-spark","messages":[{"role":"user","content":"Write a Python quicksort"}]}'

# Streaming: add "stream":true; returns standard SSE (data: {...} / data: [DONE])

3. Images (curl)

Copycurl -X POST "BASE/v1/images/generations" \
  -H "Authorization: Bearer $MUSE2API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"A Shiba Inu wearing an astronaut helmet","size":"1:1","response_format":"url"}'

# Response
{"created":1790148495,"data":[{
  "revised_prompt":"A Shiba Inu wearing an astronaut helmet",
  "url":"http://your-server:18610/v1/media/xxxx.webp",
  "kind":"image","bytes":23860}]}

# url is a full absolute URL, ready for downstream render/download; /v1/media/* needs no auth.
# For base64, pass "response_format":"b64_json".

4. Video (curl, async)

Copy# 1. Create the task
curl -X POST "BASE/v1/videos" \
  -H "Authorization: Bearer $MUSE2API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"A kitten walking across a meadow","duration":5,"size":"16:9"}'
# → {"id":"task_xxx","status":"queued"}

# 2. Poll the result
curl "BASE/v1/videos/task_xxx" -H "Authorization: Bearer $MUSE2API_KEY"
# → {"status":"succeeded","result":{"url":"/v1/media/xxx.mp4"}}

5. Python (OpenAI SDK compatible)

Copyfrom openai import OpenAI
import time

client = OpenAI(base_url="BASE/v1", api_key="m2a_...")

# Image
r = client.images.generate(model="muse-image", prompt="A Shiba Inu wearing an astronaut helmet")
print(r.data[0].url)

# Video (native endpoint, since it is async)
import requests
t = requests.post("BASE/v1/videos", json={"prompt":"Kitten walking","duration":5},
                  headers={"Authorization":"Bearer m2a_..."}).json()
while True:
    s = requests.get(f"BASE/v1/videos/{t['id']}",
                     headers={"Authorization":"Bearer m2a_..."}).json()
    if s["status"] in ("succeeded","failed"): break
    time.sleep(5)
print(s)

6. Anthropic Messages API (Claude Code & Agent SDK)

Copy# Claude Code (terminal):
export ANTHROPIC_BASE_URL="http://127.0.0.1:18610"
export ANTHROPIC_API_KEY="$MUSE2API_KEY"
claude

# Python (Anthropic SDK):
import anthropic
client = anthropic.Anthropic(base_url="http://127.0.0.1:18610", api_key="m2a_...")
msg = client.messages.create(
    model="claude-3-5-sonnet", max_tokens=1024,
    messages=[{"role": "user", "content": "Hello!"}]
)
print(msg.content[0].text)

# curl (POST /v1/messages)
curl -X POST "BASE/v1/messages" \
  -H "x-api-key: $MUSE2API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{"model":"claude-3-5-sonnet","max_tokens":1024,"messages":[{"role":"user","content":"Hello"}]}'

7. Admin API

8. Auth

9. Connecting clients / agents

Compatibility by client type