What the Genius - Song Lyrics API is and who it's for
Genius built its reputation as a crowdsourced lyrics and music annotation platform, and this API surfaces most of that data in a developer-friendly way. Rather than limiting you to a single lyrics lookup, the API gives you access to the full ecosystem: song text, artist and album metadata, community-written annotations, trending charts, web-page annotations, and detailed user contribution records.
The typical integrators are music apps that want to display synchronized or on-demand lyrics, music discovery tools that rely on trending chart data, and research or data-science projects that need large-scale access to song and artist metadata. The annotation layer also makes it interesting for anyone building educational or fan-community tools, since Genius annotations often contain rich contextual commentary that raw lyrics databases do not.
Endpoint walkthrough
The 61 endpoints are organized into logical groups. Here is what each area provides:
Search
GET /search/ performs a keyword search across all songs hosted on Genius. GET /search/multi/ extends this to all content sections on the platform, so you can surface artists, albums, and annotations in a single call rather than running separate queries.
Songs
The four song endpoints are likely the most-used entry points. GET /song/details/ returns structured metadata for a single song, including referent attachment data. GET /song/lyrics/ returns the actual lyric text. GET /song/recommendations/ surfaces related songs, useful for building discovery features without training your own model. GET /song/comments/ exposes community comment threads attached to a song.
Artists and albums
Artist endpoints cover details, discography (/artist/albums/ and /artist/songs/), activity feeds, and leaderboard position. Album endpoints mirror the pattern with details, appearances, and comments. Together, these let you reconstruct a fairly complete artist profile without scraping the Genius website.
Annotations and referents
This is where Genius differentiates itself from a plain lyrics API. A referent is a highlighted passage in a document — a song or any web page — and annotations are the explanatory notes attached to referents. GET /annotation/details/ retrieves a single annotation, while GET /annotation/versions/ shows its edit history. GET /referents/ lets you query referents by document, which means you can pull all annotated passages in a song and present them inline alongside the lyrics.
Charts
Four chart endpoints cover trending artists, albums, songs, and referents. These are useful for "what's hot" features that don't require you to build your own trending algorithm.
Web pages
GET /web-page/lookup/ and GET /web-page/referents/ expose Genius's annotation system as applied to arbitrary public web pages — not just Genius-hosted songs. If your product lets users annotate external content, this area of the API becomes relevant.
Users and contributions
Seven endpoints break down a user's contribution history into fine-grained categories: annotations, pyongs (Genius's upvote mechanic), transcriptions, suggestions, questions and answers, articles, and unreviewed annotations. These are primarily useful for applications that integrate with Genius accounts or display contributor profiles.
Videos
GET /videos/ rounds out the endpoint list, though the available data does not detail the exact response structure.
Pricing breakdown
The API uses a freemium model with four plans:
| Plan | Monthly price | Included requests | Rate limit | Overage per request |
|---|---|---|---|---|
| BASIC | $0 | 100 | — | — |
| PRO | $10 | 10,000 | 5 / second | $0.0010 |
| ULTRA | $30 | 100,000 | 5 / second | $0.0010 |
| MEGA | $200 | 1,000,000 | 5 / second | $0.0010 |
The free BASIC tier allocates only 100 requests per month, which is enough for prototyping and exploring endpoint shapes but not for any production feature with real users. At 5 requests per second, the rate limit on paid plans is the same across PRO, ULTRA, and MEGA — the differentiator is purely the included request budget. The $0.001 overage charge on paid plans is predictable: 10,000 extra requests costs $10, which maps neatly onto the PRO tier cost, so heavy overage essentially prices you into the next plan tier automatically.
For a music app showing lyrics on demand, PRO's 10,000 monthly requests supports roughly 10,000 lyric page views (assuming one request per view). ULTRA at $30 handles 100,000 views, and MEGA at $200 handles one million. Those are reasonable unit economics for an ad-supported or subscription music product.
Practical use cases
Lyrics display: The most straightforward use is fetching lyrics via /song/lyrics/ and displaying them inside an app. Pair it with /song/details/ to get metadata for proper attribution.
Music discovery: Combine /chart/songs/ with /song/recommendations/ to build a lightweight recommendation and trending surface without any machine learning infrastructure on your side.
Annotation viewers: Pull referents for a song and overlay them on the lyrics so users can tap highlighted lines to read community commentary — a feature that directly replicates the core Genius experience inside a third-party app.
Artist profile pages: Chain /artist/details/, /artist/albums/, and /artist/songs/ to assemble a complete artist page. The activity and leaderboard endpoints add social context if that fits your product.
Research and data pipelines: The broad search and multi-search endpoints, combined with overage pricing that scales linearly, make bulk data collection viable on MEGA or a carefully managed ULTRA plan.
Limitations and things to check before integrating
Low free-tier quota: 100 requests per month is genuinely minimal. Plan to move to PRO quickly once you move beyond local development.
Uniform rate limit: All paid tiers share a 5 requests/second ceiling. If your use case involves burst traffic — for example, a playlist that pre-fetches lyrics for 20 queued songs simultaneously — you will need to implement client-side queuing.
Latency: The average response time is 830 ms. For interactive features like a lyrics view that opens when a song starts playing, you should pre-fetch or cache aggressively, since 830 ms is perceptible in a UI context.
Endpoint count vs. documentation depth: The fact sheet lists 61 endpoints but only details 40. Verify the full endpoint surface and request/response schemas in the marketplace documentation before planning an integration that depends on the less-documented endpoints.
Terms of use: Genius's own terms around lyrics reproduction apply regardless of which API tier you use. Confirm that your intended use case — particularly any public display of full lyrics — is permitted under both the marketplace terms and Genius's own licensing policies.
Getting started
Subscribe through the API marketplace (over 8,000 developers already have) and start with the BASIC plan to validate your integration. Use /search/ or /search/multi/ as your entry point to resolve song or artist identifiers, then follow up with the specific detail or lyrics endpoints. Given the 830 ms average latency, implement response caching from day one rather than treating it as an optimization for later. Once you have a clear picture of your request volume per user session, you can choose the right paid tier before launching to production.
