Luma API

TTAPI Luma API, text to video service

Generate Video

POST https://api.ttapi.io/luma/v1/generations

Generate video from text & images. Task results is async

💡

Generally speaking, fast mode will return within 300 seconds, but this also depends on the complexity of the prompt and the official timeliness. relax mode usually cannot guarantee timeliness and generally returns within 1 minute to 30 minutes.

Headers

Name	Type	Description
TT-API-KEY*	String	Your API Key in TT API used for request authorization

Request Body

Name	Type	Required	Description
userPrompt	String	true	Generate video text prompts, `for example: a red car driving on a road, for details`, Note: Luma officially supports multiple languages, but after trying, English is still the best
~~enhancePrompt~~	Boolean	false	~~If you are not sure how to write a prompt, use `true` for the 'enhancePrompt' to let AI turn your simple concept into a detailed prompt. the default value is `true`, Please refer to~~
imageUrl	String	false	Set the image link of the starting frames of the video. If this parameter is used, the starting frame of the video will be the image of this image. Please refer to
imageEndUrl	String	false	Set the image link of the ending frames of the video. If this parameter is used, the ending frame of the video will be the image of this image. Please refer to
hookUrl	String	false	Send a request to the address for task completion or failed notification. If not set you need to request fetch endpoint to get response
loop	Boolean	false	Whether to generate a looping video, which can be understood as the start frame and the end frame are the same. The optional value is `false` or `true`. The default value is `false`
aspectRatio	String	false	Video ratio Support: `9:16` `3:4` `1:1` `4:3` `16:9` `21:9` The default is `9:16`
modelName	String	false	Use Mode Support `ray-v1` `ray-v2` `ray-v2-flash` The default is `ray-v2`
duration	String	false	Video Length Support `5s` `10s` The default is `5s`, 10s long videos are only available in 720p videos
resolution	String	false	Video clarity Support is `720p` `1080p` The default is `720p`
useMode	String	false	Generate speed mode Support is `relax` `fast` The default is `relax`. Different modes use different quotas, detail see

Example Request

import requests
endpoint = "https://api.ttapi.io/luma/v1/generations"
headers = {
    "TT-API-KEY": your_key
}
data = {
    "userPrompt": "a red car driving on a road",
}
response = requests.post(endpoint, headers=headers, json=data)
print(response.status_code)
print(response.json())

Example Response

{
    "status": "SUCCESS",
    "message": "",
    "data": {
        "jobId": "ed1a1b01-7d64-4c8a-acaa-71185d23a2f3"
    }
}

Extend Video

POST https://api.ttapi.io/luma/v1/extend

Generate a new video based on the previous video. Task results is async

Headers

Name	Type	Description
TT-API-KEY*	String	Your API Key in TT API used for request authorization

Request Body

Name	Type	Required	Description
jobId	String	true	The jobId of the previous video
userPrompt	String	true	Generate video text prompts, `for example: a red car driving on a road, for details`, Note: Luma officially supports multiple languages, but after trying, English is still the best
~~enhancePrompt~~	Boolean	false	~~If you are not sure how to write a prompt, use `true` for the 'enhancePrompt' to let AI turn your simple concept into a detailed prompt. the default value is `true`~~
~~lastVideoImage~~	Boolean	false	~~Whether the current video references the last frame of the previous video. The optional value is `false` or `true`. The default value is `true`~~
hookUrl	String	false	Send a request to the address for task completion or failed notification. If not set you need to request fetch endpoint to get response
loop	Boolean	false	Whether to generate a looping video, which can be understood as the start frame and the end frame are the same. The optional value is `false` or `true`. The default value is `false`

Example Request

import requests
endpoint = "https://api.ttapi.io/luma/v1/extend"
headers = {
    "TT-API-KEY": your_key
}
data = {
    "jobId": "ed1a1b01-7d64-4c8a-acaa-71185d23a2f3",
    "userPrompt": "a red car driving on a road",
}
response = requests.post(endpoint, headers=headers, json=data)
print(response.status_code)
print(response.json())

Example Response

{
    "status": "SUCCESS",
    "message": "",
    "data": {
        "jobId": "ed1a1b01-7d64-4c8a-acaa-71185d23a2f3"
    }
}

Fetch Video

POST/GET https://api.ttapi.io/luma/v1/fetch

Get the generate video task result query, the returned json data is same with the return in hookUrl

Headers

Name	Type	Description
TT-API-KEY*	String	Your API Key in TT API used for request authorization

Request Body / Query param

Name	Type	Required	Description
jobId	String	true	ed1a1b01-7d64-4c8a-acaa-71185d23a2f3

Example Request

import requests
endpoint = "https://api.ttapi.io/luma/v1/fetch"
headers = {
    "TT-API-KEY": your_key
}
data = {
    "jobId": "ed1a1b01-7d64-4c8a-acaa-71185d23a2f3",
}
response = requests.post(endpoint, headers=headers, json=data)
print(response.status_code)
print(response.json())

Async callback & fetch json

{
    "status": "SUCCESS",
    "message": "success",
    "jobId": "6eb34b44-64c8-4629-a0a1-608737de9583",
    "data": {
        "jobId": "6eb34b44-64c8-4629-a0a1-608737de9583",
        "userPrompt": "Sanrio style illustration of Hello Kitty dressed ",
        "width": "512",
        "height": "512",
        "imageUrl": null,
        "imageEndUrl": null,
        "hookUrl": "https://webhook-test.com/72b9baa490830671b5cd068815788b7e",
        "videoUrl": "https://storage.cdn-luma.com/lit_lite_inference_im2vid_v1.0/11be9408-f349-46df-a43e-a3b57002c1cd/watermarked_video09312742f491e41c19974b52babdba615.mp4",
        "quota": 25
    }

Swap Face LLM