Evil0ctal

Douyin_TikTok_Download_API: Collect Data and Download Videos

A self-hosted API and console for collecting Douyin and TikTok posts, profiles, comments, and playlists, with no-watermark media downloads. It offers REST, MCP, and CLI access, and stores collected data in PostgreSQL.

PythonApache-2.0Self HostedAPIWeb Scraping
View on GitHub Compare
Stars
20.4k
Forks
2.8k
Stars, last 8 days
+154

Our take

Health score

54/100

How it is scored
Activity
75
Community
20
Issues
53
Pull requests
69

+154 stars in the last 8 days

  • Slower to close issues than 91% of the projects we track

Overview

Douyin_TikTok_Download_API is a self-hosted data collection service for Douyin and TikTok. It provides one place to retrieve platform content, download media, and keep an archive in your own database instead of relying on a hosted scraping service.

The project is aimed at developers and operators who want to run collection infrastructure themselves. REST, MCP, and CLI interfaces share the same service layer, while a web console provides controls for identities, scheduled tasks, archived content, and diagnostics.

Key Features

  • Fetch posts, profiles, author posts, liked posts, playlists, comments, and comment replies from both platforms.
  • Download videos and image albums without watermarks by selecting the clean stream provided by the platform.
  • Expose the service through a REST API, an MCP server, and a command-line interface.
  • Archive collected data in PostgreSQL, with Redis used alongside it.
  • Manage identities, request scheduling, API keys, downloads, and diagnostics through a web console.
  • Collect content asynchronously by default, with an option to wait for a synchronous response.
  • Deploy the service and its supporting components with Docker Compose.

Use Cases

  • Developers can build applications that retrieve Douyin or TikTok content through a self-hosted API.
  • Researchers and analysts can collect posts, profiles, and comments while keeping the resulting archive under their control.
  • Teams can use the MCP interface to make supported collection operations available to compatible AI clients.
  • Individuals can download videos or image albums and organize media locally, including bulk author downloads.
  • Operators can monitor creators or posts and schedule recurring collection on their own infrastructure.

Getting Started

The recommended deployment uses Docker Compose. After cloning the repository and configuring the required secrets and database URLs, start the stack with:

docker compose -p dtk -f docker/compose.yml up -d

See the README and installation guide for setup requirements, configuration, and operations.

Considerations

  • This is infrastructure to run and maintain yourself, not a hosted data service. Docker Compose is the recommended deployment path, and the stack needs PostgreSQL with TimescaleDB and Redis.
  • Platform access can be affected by identity validity, rate limits, and changes to platform endpoints or signing behavior. The project includes identity health and request scheduling mechanisms, but collection still depends on upstream services.
  • Douyin follower and following lists are not supported. TikTok supports those endpoints; the README says Douyin restricts them to signed-in sessions.
  • The current main branch is v5 and does not include the Bilibili support available in v4.
  • Users are responsible for respecting platform terms, applicable law, and the rights of people whose content they collect.

Found this useful?

Share it with someone who would like Douyin_TikTok_Download_API.