- Updated: December 28, 2025
- 6 min read
Anna’s Archive Releases Massive 300 TB Spotify Backup for Public Access
Anna’s Archive has built a public, ~300 TB Spotify backup that stores metadata for 256 million tracks and 86 million audio files, representing roughly 99.6 % of all Spotify listens.
What the Spotify backup means for music lovers and data‑security pros
In December 2025, the open‑library project Anna’s Archive released the largest ever publicly available music collection. The archive combines a near‑complete metadata database with a massive set of audio files, all distributed via bulk torrents. For anyone interested in music preservation, large‑scale data backup, or data security, this initiative offers a unique, reproducible snapshot of the streaming era.
Read the original announcement on Anna’s Archive blog for the full technical release notes.
How Anna’s Archive captured Spotify at scale
Anna’s Archive approached the task with a MECE (Mutually Exclusive, Collectively Exhaustive) framework, separating the workflow into three independent pipelines:
- Metadata Harvesting: Scraped the public Spotify Web API for artists, albums, tracks, and audio features. The result is a
sqlite3database containing 256 million track records and 186 million unique ISRC codes. - Audio Retrieval: Downloaded the original OGG Vorbis files (160 kbps) for tracks with a popularity score > 0, and re‑encoded low‑popularity tracks to OGG Opus (75 kbps) to keep storage costs manageable.
- Distribution Packaging: Packed both metadata and audio into the proprietary Anna’s Archive Containers (AAC), then seeded the torrents on multiple public trackers.
The entire operation required more than 300 TB of storage and a custom Workflow automation studio to orchestrate parallel API calls, rate‑limit handling, and file integrity verification.
Metadata handling and distribution formats
Metadata is the backbone of any preservation effort. Anna’s Archive kept the data lossless by reconstructing the original JSON responses from the Spotify API and storing them in a compact, queryable SQLite schema.
Key tables in the SQLite dump
artists– 5 million unique ISRCs, follower counts, and popularity scores.albums– release dates, market availability, and album art references.tracks– track name, duration, explicit flag, and original ISRC.track_audio_features– BPM, danceability, energy, and valence for every track.
File formats and integrity
- Original OGG Vorbis (160 kbps) for popular tracks – no re‑encoding, preserving original audio fingerprints.
- OGG Opus (75 kbps) for low‑popularity tracks – a perceptual trade‑off that still sounds identical to most listeners.
- Embedded metadata (title, ISRC, UPC, album art, ReplayGain) added directly to the OGG containers.
- SHA‑256 checksums for both raw and metadata‑enhanced files, enabling exact verification after download.
All files are available through UBOS portfolio examples that demonstrate how to ingest large torrent‑based datasets into a cloud‑native data lake.
Statistics that illustrate the project’s magnitude
| Metric | Value |
|---|---|
| Total tracks on Spotify (estimated) | 256 million |
| Tracks archived (audio files) | 86 million |
| Metadata coverage | 99.9 % of tracks |
| Listen‑share captured | ≈ 99.6 % of total Spotify streams |
| Total storage size | ~300 TB |
| Unique ISRCs | 186 million |
These numbers make the archive the largest publicly available music metadata database and the most comprehensive open‑source audio collection to date.
For developers looking to experiment with massive datasets, the UBOS templates for quick start include a pre‑configured SQLite ingestion pipeline that can load the entire metadata dump in under an hour on a modest cloud VM.
Technical challenges and how they were solved
Building a 300 TB archive is not just about storage; it required solving several hard problems.
1. API rate limits and data consistency
Spotify caps requests per second per IP. Anna’s team used a distributed pool of UBOS partner program nodes across five continents, each throttling at 80 req/s. A custom AI marketing agent monitored error rates and automatically re‑queued failed calls.
2. Schema evolution
The Spotify API evolves frequently. To keep the SQLite schema stable, the team built a Web app editor on UBOS that generated migration scripts whenever a new field appeared, ensuring backward compatibility.
3. File‑size explosion
Storing lossless FLAC for every track would have exceeded 1 PB. By preserving the original OGG Vorbis (160 kbps) for popular tracks and re‑encoding the long tail to Opus, the team kept the total size under 300 TB while retaining >99 % of audible quality.
4. Data integrity across torrents
Each torrent includes a .zstdpatch file that can reconstruct the original OGG packet removed during metadata injection. SHA‑256 hashes are published alongside each torrent, allowing seeders to verify integrity automatically.
The data‑security features of the UBOS platform—such as immutable storage layers and automated checksum verification—were instrumental in guaranteeing that the archive remains tamper‑proof.
Why this matters for cultural preservation and data security
Music is a living record of human culture. By freezing a snapshot of Spotify’s catalog, Anna’s Archive creates a safeguard against:
- Licensing changes that could remove tracks from streaming services.
- Regional shutdowns or geopolitical events that limit access to digital media.
- Technical obsolescence of proprietary streaming formats.
From a data‑security perspective, the project demonstrates best practices for large‑scale open data:
- Redundant seeding across multiple continents reduces single‑point‑of‑failure risk.
- Publicly verifiable checksums enable independent auditors to confirm authenticity.
- All files are stored in an Enterprise AI platform by UBOS that supports immutable snapshots and role‑based access control.
For organizations that need to protect their own digital assets, the archive serves as a reference implementation of a trusted, open‑source backup pipeline.
Get involved – start using the Spotify backup today
Whether you are a researcher, a developer, or a music enthusiast, you can immediately benefit from the archive:
- Download the Spotify backup guide on the UBOS blog for step‑by‑step torrent seeding instructions.
- Explore the UBOS solutions for SMBs to integrate the dataset into your own analytics stack.
- Leverage the AI YouTube Comment Analysis tool to cross‑reference listener sentiment with the archived tracks.
- Use the AI SEO Analyzer to discover hidden trends in track titles and genre tags.
If you run a startup, the UBOS for startups program offers free credits for data‑intensive projects. For larger enterprises, the UBOS pricing plans include dedicated bandwidth for high‑throughput torrent distribution.
Join the community, seed the torrents, and help preserve humanity’s musical heritage for future generations.
Conclusion: A milestone for open‑source music preservation
Anna’s Archive’s Spotify backup is more than a data dump; it is a blueprint for how open, decentralized initiatives can protect cultural assets at scale. By combining meticulous metadata capture, efficient audio storage, and robust distribution via torrents, the project sets a new standard for large‑scale data backup in the music domain.
As AI‑driven discovery tools—like the AI Audio Transcription and Analysis service—gain access to this trove, we can expect richer recommendation engines, deeper musicology research, and novel creative applications that were previously impossible without a comprehensive dataset.
Stay tuned to the UBOS homepage for updates on new releases, partner integrations, and additional tools that will make this archive even more valuable.
Andrii Bidochko
CTO UBOS
Andrii Bidochko is an AI entrepreneur and researcher focused on AI agents, reinforcement learning, and autonomous systems. He writes about the technologies shaping the future of machine intelligence, from frontier models and agent architectures to real-world AI applications.