ElevenLabs Music API by ElevenLabs

# Fix list: ElevenLabs Music API

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/elevenlabs-music, the October 2026 research run, assessed 1 October 2026. Grade B, 67 out of 100.

This is everything the published grade says the listing lacks, the biggest possible gain to the total first. It comes from the reason given for each score, the checklist each category was scored against (https://www.anchorterminal.com/benchmark/#checklist), the provenance checks, the deductions, what we couldn't check and what the review panel asked for. A fix counts at the next check, once it's public.

For a coding agent working on ElevenLabs Music API: work through the items below in the product, its docs and its public pages. Each category gives the reason for its score, with the points each checklist item earned, and the checklist itself, so the gap is the items that earned less than their points. Change the product, not the wording, and keep a note of what you changed and where it's published.

## 1. Payments & pricing, 20 out of 100, up to 10 more on the total

Why it scored 20: No x402, MPP or L402 (0). $0.15 a minute of music published without a login, plus plan prices (20). The Free plan's 3 minutes are web app only, since the Music API is disabled on Free until you top up (0). Pay-as-you-go needs a person to add a card and at least $5 (0).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-payments):

The published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).

- 40, a machine payment protocol (x402, MPP or L402) on the tool's own endpoints. 10 to 30 when it covers only some endpoints or only goes through a third party, and the note says which.
- 20, per-call or per-unit pricing published without a login. 10 for public plan-only pricing, 0 for "contact sales" or prices behind a login.
- 20, a free tier or trial that doesn't need a card.
- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).

Payment platforms and agent wallets rarely charge for their own API over a machine protocol, so the first line has steps for them, and the highest one that applies counts. 40 when x402, MPP or L402 runs on all their own endpoints, 30 when it runs on part of their own API, 25 when their merchants can accept one, 20 for running a facilitator, 15 for paying as a buyer, and 0 when the only protocol is their own. Merchant acceptance sits above a facilitator because the platform's own customers can charge agents through it, while a facilitator settles for sellers who wire up the protocol themselves. The counter-argument (a facilitator does more for the protocol as a whole) has a point. Each note says which step applied.

Open-source software you run yourself is scored on its hosted or paid option if it has one. A free, self-hosted package with nothing to buy gets 20, 20 and 20 for the last three lines, and 0 to 40 for the first only if it ships a payment protocol.

## 2. Reliability, 63 out of 100, up to 7.4 more on the total

Why it scored 63: Status page on incident.io with component history for ElevenAPI, ElevenCreative and others, though Music isn't its own component (20). One long incident in the last 90 days touched Music. On 2026-08-24 media loading failed across Studio, Music, Dubbing and other products for about ten hours, with the API degraded, plus a 9-minute US error spike on 2026-07-08 (10). Concurrency limits published per plan, 2 on Free to 15 on Scale and Business (15). The 429 page names `too_many_concurrent_requests` and `system_busy` and says to retry the second, and the API error reference asks for exponential backoff, with no Retry-After header (8). No idempotency key or safe-retry guidance for generation (0). No SLA published, only custom SLAs for enterprise (0). The Music API is generally available (10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability):

Hosted APIs, MCP servers, models and platforms.

- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).
- 0 to 30, the incident record for the last 90 days on that page. 30 for a clean record or trivial incidents only, 20 for minor incidents only, 10 for one major outage (an hour or more of a core API down, or errors across the board), 0 for several. 5 when there's no history we could read, and the note says so.
- 15, rate limits documented with numbers.
- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.
- 10, an SLA published for any paid tier.
- 10, the surface agents use is generally available, not beta or preview.

Local packages, SDKs, frameworks and stdio MCP servers.

- 20, installs from an official package with supported runtimes stated.
- 25, a public CI and test suite, passing on the default branch.
- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).
- 15, semver discipline and breaking changes called out in a changelog.
- 15, version 1.0 or later, or declared stable.

Protocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors.

## 3. Agent ergonomics, 65 out of 100, up to 5.7 more on the total

Why it scored 65: Audio comes back as a file in the response body with a `song-id` header, not base64 in JSON (15), and the caller sets both length and format (10). A streaming endpoint for music, no webhook, and we didn't find paginated history for music generations (10). Validation errors return 422 with detail and the two 429 types carry named codes (15). No idempotency key and no statement on whether failed generations are charged (0). Only `prompt` is needed and everything else has a default (7), official SDKs for Python and TypeScript (8).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics):

- 0 to 25, context cost. For MCP, the number and size of the tool definitions (25 for ten or fewer compact tools, 15 for 11 to 30, 5 for more than 30, plus up to 10 back for toolsets, dynamic loading or read-only subsets). For APIs, whether responses can be sized (field selection, limits, summaries).
- 20, pagination, filtering and output-size controls.
- 20, actionable, documented error responses, codes and messages an agent can recover from.
- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.
- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.

Models are read for tool use, structured output, prompt caching, context length, batch and SDKs. Frameworks for how much code and how many defaults a tool-calling agent with MCP needs.

## 4. Security & auth, 77 out of 100, up to 4 more on the total

Why it scored 77: API keys can be limited to chosen endpoints, given a credit cap, an expiry from 15 minutes to 30 days and an IP allowlist, and service-account keys exist for production. Keys found in public GitHub repos are disabled automatically (30). Endpoint-scoped keys give least privilege. Music has no destructive actions to confirm (15). Returns generated audio, no untrusted content (10). Credit caps show spend per key, and workspace audit logs exist but only on Enterprise (10). security.txt valid to 2027-03-01 with a vulnerability disclosure programme (5), no paid bug bounty found (0), SOC 2 Type II announced by ElevenLabs on X, with a trust centre at compliance.elevenlabs.io that didn't render for us (7), no public advisories found (0).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-security):

- 0 to 30, the credential model. 30 for OAuth 2.1 with scopes, or scoped and revocable keys with rotation. 20 for plain revocable API keys. 10 for one all-powerful key. 10 off when a secret can travel in a URL query string as a documented option.
- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.
- 0 to 15, prompt-injection posture where the tool returns untrusted content (documented mitigations or guidance). A tool that returns no untrusted content gets 10.
- 0 to 15, audit logs or per-call visibility for the operator.
- 0 to 20, a security programme. security.txt or a disclosure policy, a bug bounty, SOC 2 or ISO 27001, advisories handled in public.

Models are read for retention, whether API data trains models (and whether that's off by default), zero-retention options and certifications. Frameworks for telemetry defaults, approval hooks, guardrails and sandboxing.

## 5. Schema & documentation, 86 out of 100, up to 2.3 more on the total

Why it scored 86: Public OpenAPI document at api.elevenlabs.io/openapi.json (25). llms.txt with Markdown pages for the music quickstart, streaming, composition plans and inpainting (10). Compose, stems and inpainting are described with purpose and when to use a composition plan instead of a prompt, but the overview says tracks run to 5 minutes while the reference accepts 600,000 ms (12). Typed inputs with enums for `model_id` and `output_format`, ranges for `music_length_ms` (3,000 to 600,000) and `seed`, a 4,100-character prompt cap and prompt and plan marked mutually exclusive (14). Examples in several languages, but only a generic 422 is documented per endpoint (10). Versioned model IDs and a dated weekly changelog (15).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-schema):

APIs and MCP servers.

- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).
- 10, llms.txt or Markdown docs served for agents.
- 0 to 20, descriptions that say what a tool is for, when to use it and when not to, read from the tool definitions in the source or the API reference.
- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.
- 0 to 15, examples and documented error responses.
- 15, versioning and a public changelog.

Models are read from the API reference, the OpenAPI file, llms.txt, the structured-output and tool-use docs and the model cards. Frameworks from docs a model can follow, typed interfaces, examples and the API reference.

## 6. Maintenance & community, 80 out of 100, up to 1.8 more on the total

Why it scored 80: Changelog entries on 2026-09-28, with Music changes on 2026-09-07 (`music_v2_5` in the API and SDKs) and 2026-09-14 (30). Ten dated changelog entries between 2026-08-17 and 2026-09-28 (20). Public weekly changelog and a help centre, but we didn't check how fast GitHub issues get answers (10). Python SDK 2.70.0 released on 2026-09-28 after eight releases since 2026-08-11, and the npm package is on 2.70.0 too (15). Packages are current, CI not checked (5).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance):

- 0 to 30, time since the last release, or the last published model or API change for a closed service. 30 within 30 days, 20 within 90, 10 within 180, 0 older.
- 20, at least three releases or dated changelog entries in the last 90 days.
- 0 to 25, responsiveness. Issues and pull requests answered on GitHub (the open issues and how recent the replies are). For closed services, a public changelog and a support or community channel that answers, 0 to 15.
- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).
- 10, package health, current dependencies and CI.

Models are read for deprecation notice periods and model churn rather than release counts.

## 7. Transparency & trust, 79 out of 100, up to 1.8 more on the total

Made of editorial 65, provenance 92.

Why it scored 79: Closed service with general terms plus Music Terms updated 2026-05-26 that spell out what output may be used for. The SDKs are MIT (15). Privacy policy, DPA and a Zero Retention Mode exist, but Zero Retention is enterprise only and the docs list Music as not yet supported, so music prompts and output are retained by default (20). Dated deprecation notices in the changelog, such as the local MCP server on 2026-08-22, but `music_v1` is called legacy without a date (15). Regional servers for the US, EU, India and Singapore and a DPA, subprocessor list not read (15).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-transparency):

- 0 to 30, source availability and licence clarity. 30 for open source under an OSI licence, 15 for closed with clear terms, 0 for unclear terms.
- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).
- 0 to 20, a deprecation policy or notices with dates.
- 0 to 20, telemetry disclosed with an opt-out (local software), or subprocessors and data locations disclosed (hosted).

The other half of Transparency and trust is the provenance score, computed from checked facts (below). The category score is the mean of the two.

Provenance checks not met in full (half of this category, computed from checked facts):

- Domain age: elevenlabs.io, registered 2021-12-15 (4 years) (7 of 15)

## What we couldn't check

What we couldn't read counted as absent. Publishing it on a page a plain HTTP fetch can read (not only in a browser) lets the next check count it.

- The help centre lists concurrency of 2 (Free) to 15 (Scale and Business), while the listing had 6 to 40. The pricing page only gives Speech Engine concurrency (4 to 40). We used the help centre figures.
- The hosted MCP docs list agent and speech tools only, and elevenlabs.io/mcp says ElevenCreative tools, music among them, are coming soon. We removed the MCP transports and registry name from this music listing until music tools appear there.
- Whether failed music generations are charged isn't stated.
- We couldn't read the subprocessor list or bug bounty terms behind the trust centre.

## Weaknesses

- Music API is disabled on the Free plan and pay-as-you-go needs a card
- Music Terms exclude customers in political, religious, adult, tobacco, weapons and prescription drug sectors, and prompts can't name artists or songs
- Zero Retention Mode doesn't cover Music yet
- No idempotency key, no SLA and no music-specific rate limit
- Docs disagree on maximum length (5 minutes in the overview, 600,000 ms in the API reference)

## What costs an agent a turn today

The notes we give agents before they call it. Each one is a workaround an agent shouldn't need.

- Strip artist and song names from user prompts before calling, or the request breaks the Music Terms
- Pin `model_id`. The API default is still `music_v1`
- Set `music_length_ms` and `force_instrumental: true` for background beds. Billing is by minute of audio
- On 429, retry `system_busy` with exponential backoff, but treat `too_many_concurrent_requests` as a plan limit and queue
- Don't plan on the MCP server for music. Its documented tools cover agents and speech

## What the review panel asked for

- Music on hosted MCP
- Reconcile the length limit
- State failed-generation billing

## When it's done

Send what changed and where it's published as a dispute (https://www.anchorterminal.com/builders/#disputes, or `POST https://www.anchorterminal.com/api/v1/contact` with `"kind": "dispute"`). Disputes are answered in public, and the listing is checked again by the same checklist. Paying for an audit or a listing claim changes nothing here.