OpenAI Sora 2 Pro Characters API: AI Text Generation

Sora 2 Pro Characters API Reference

Endpoint

POST https://api.muapi.ai/api/v1/openai-sora-2-pro-characters

Submit a job with your MuApi API key in the x-api-key header, then poll https://api.muapi.ai/api/v1/predictions/{request_id}/result until status is completed.

Parameters

Name	Type	Required	Description
origin_task_id	string	Yes	The task ID of the original Sora video.
character_user_name	string	No	A globally-unique handle for your character. Reference it in later prompts by prefixing it with @ (e.g., @sky_ranger).
prompt	string	Yes	In one short line, state stable traits (e.g., “cheerful barista, green apron, warm smile”); avoid camera directions, contradictions, or disallowed celebrity likeness.
safety_instruction	string	No	Briefly list any boundaries (“no violence, politics, or alcohol; PG-13 max”); tighter wording helps the model enforce your content limits.
start_time	float	Yes	Enter the clip's start and end times in seconds. The segment you select must be completely inside the original video and between 1 s–4 s long; that slice will be used as the character's training material.
end_time	float	Yes	Enter the clip's start and end times in seconds. The segment you select must be completely inside the original video and between 1 s–4 s long; that slice will be used as the character's training material.

cURL example

REQUEST_ID=$(curl -s -X POST https://api.muapi.ai/api/v1/openai-sora-2-pro-characters \
  -H "x-api-key: $MUAPI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"origin_task_id":"<origin_task_id>","prompt":"<prompt>","start_time":"<start_time>","end_time":"<end_time>"}' | jq -r .request_id)

curl -s https://api.muapi.ai/api/v1/predictions/$REQUEST_ID/result -H "x-api-key: $MUAPI_API_KEY"

Python example

import os, time, requests

API = "https://api.muapi.ai/api/v1"
headers = {"x-api-key": os.environ["MUAPI_API_KEY"]}

r = requests.post(f"{API}/openai-sora-2-pro-characters", headers=headers, json={"origin_task_id":"<origin_task_id>","prompt":"<prompt>","start_time":"<start_time>","end_time":"<end_time>"})
request_id = r.json()["request_id"]

while True:
    res = requests.get(f"{API}/predictions/{request_id}/result", headers=headers).json()
    if res["status"] == "completed":
        print(res["outputs"]); break
    if res["status"] == "failed":
        raise RuntimeError(res.get("error"))
    time.sleep(3)

Full docs and agent/MCP integration: llms.txt. Get an API key at muapi.ai/access-keys.