Luma API

中文

Using Golang, convert the video capabilities of the LumaAI Dream Machine into an API format for calls, making it easier to integrate quickly into third-party systems.

Disclaimer

• This project is released on GitHub under the MIT license, free and open-source for educational purposes.

Good news

I provide the Luma AI API, no deployment is required, no subscription to Luma is required. Lower price, more convenient to use the Luma API.

Website: https://api.bltcy.ai

Supported Features

• Luma API supports video generation and allows uploading reference images.
• Users can upload reference images in advance or provide third-party image links or base64 encoded images when submitting video tasks.
• API interface security verification.
• Simplified deployment process, supporting both docker-compose and docker.
• Provides standardized services compatible with OpenAI's interface format, supporting both streaming and non-streaming output.
• Supports customization of the OpenAI Chat response format based on Go Template syntax.
• Compatible with frontend projects like chat-next-web.
• Automatic keep-alive (max keep-alive 7 days).
• [] Adapts to intermediary projects like New API.
• [] Pre-detection of sensitive words.

Supported Endpoints

Video generation, supports uploading two images (start and end). [x] /generations Video extension, extends by 5 seconds each time. [x] /generations/:task_id/extend Query single video task. [x] /generations/:task_id Pre-upload images. [x] /generations/file_upload Get download URL without watermark. [x] /generations/:task_id/download_video_url Get account subscription information (balance, etc.). [x] /subscription/usage Get account information. [x] /users/me

Web UI

https://github.com/Dooy/chatgpt-web-midjourney-proxy
Online
https://api-chat.gptbest.vip

API Docs

http://localhost:8000/swagger/index.html

Request example

curl --location 'http://localhost:8000/luma/generations/' \
--header 'Authorization: Bearer {SECRET_TOKEN}' \
--header 'Content-Type: application/json' \
--data '{
    "user_prompt": "Increase dynamic",
    "aspect_ratio": "16:9",
    "expand_prompt": true,
    "image_url": "https://api.bltcy.ai/logo.png" 
}'

Response example

{
    "created_at": "2024-06-18T14:47:43.318498Z",
    "estimate_wait_seconds": null,
    "id": "a4a8f1ff-e59b-4969-abcb-98b3e16f003e",
    "liked": null,
    "prompt": "Increase dynamic",
    "state": "pending",
    "video": null
}

Deployment

Configuration

Log in using Google Chrome https://lumalabs.ai/dream-machine/ Get cookie

Env Environment Variables

env	description	default value
COOKIE	Cookie obtained from the image above for the Luma account	None
BASE_URL	Request API URL for Luma	https://internal-api.virginia.labs.lumalabs.ai
PROXY	HTTP proxy	None
PORT	Port	8000
SECRET_TOKEN	Luma API seurity http header Bearer token	None
ROTATE_LOGS	Whether to rotate logs daily	Yes
LOG_DIR	Log output path	./logs
DEBUG	Whether to enable Debug logging	No
PPROF	Whether to enable Pprof performance analysis, uses port 8005 if enabled	No
CHAT_TIME_OUT	Chat request time out	600 s
CHAT_TEMPLATE_DIR	Chat template path	./template

Docker Deployment

Step-by-step guide on how to run a Docker container with specific environment variables and port mappings. Sensitive details like SQL names, passwords, and IP addresses are replaced with placeholders for this guide.

docker run --name luma-api -d -p 8000:8000 \
-e COOKIE=xxxx  \
 lumaapi/luma-api

docker-compose deployment

docker-compose pull && docker-compose up -d

docker-compose.yml

version: '3.2'

services:
  sunoapi:
    image: lumaapi/luma-api:latest
    container_name: luma-api
    restart: always
    ports:
      - "8000:8000"
    volumes:
      - ./logs:/logs
    environment:
      - COOKIE=xxxx

Local Deployment

• Golang 1.20+

git clone https://github.com/LumaAI-API/Luma-API.git
cd Luma-API
go mod tidy

License

MIT © Luma API

Buy Me a Coke

This project is open-source on GitHub under the MIT license and is free of charge. If you find this project helpful, please give it a star and share it. Thank you for your support!