Smart search

ArchiveTube’s search feature can be enhanced by embedding a video’s title and description. This is done using qwen/qwen3-embedding-8b through OpenRouter. When using smart search, each search query gets embedded server-side; this could be expensive if you have a public instance.

Warning

If you already have archived videos and smart search was disabled, you need to follow the migration guide.

It supports any OpenRouter-like API, including HackClub AI. Add this to your config, replace the apikey, and restart your container.

If you want to use OpenRouter instead, change the backend to https://openrouter.ai/api/v1/embeddings.

[smart_search]
enabled = true
backend = "https://ai.hackclub.com/proxy/v1/embeddings"
model = "qwen/qwen3-embedding-8b"
apikey = "sk-hc-v1-"

Migration

If you already have archived videos that were archived without smart search enabled, you’ll need to embed them.

1. Clone ArchiveTube on your PC.
2. Open the embedding-migration folder.
3. Stop your container.
4. Back up the database.
5. Copy the database file to ./embedding-migration/database.db.
6. Create an apikey file with your HackClub AI key inside.
7. Install requirements with pip install -r requirements.txt.
8. Run python main.py.
9. Copy database.db back to the database.

This will embed ALL titles and descriptions. This can be expensive if you have a lot of archived videos.