ElevenLabs Voice Cloning and Voice Design API by ElevenLabs

# Fix list: ElevenLabs Voice Cloning and Voice Design API

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/elevenlabs-voice-cloning, the October 2026 research run, assessed 1 October 2026. Grade BB, 73.8 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 Voice Cloning and Voice Design 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, 30 out of 100, up to 8.8 more on the total

Why it scored 30: No x402, MPP or L402 (0). Per-unit prices for speech from a clone are public, such as $0.022 per 1,000 characters for TTS v4 on the API pricing page (20). The Free plan needs no card and includes Voice Design, but no cloning (10 of 20). Signup is a human browser flow (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 at status.elevenlabs.io (incident.io) with component history (20). The last 90 days hold one major incident on the surface a clone is used through, elevated error rates on the TTS and STT APIs for about 2 hours on 3 August, plus about 8 hours of degraded TTS latency on 26 August and a long list of incidents on other products (10). Concurrency per plan is published, 2 on Free up to 15 on Scale and Business (15). The 429 codes `too_many_concurrent_requests` and `system_busy` are documented with exponential backoff guidance (8). No idempotency key or safe-retry guidance for creating a voice (0). Enterprise mentions custom SLAs, but no SLA is published (0). IVC, PVC and Voice Design are 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. Security & auth, 84 out of 100, up to 2.8 more on the total

Why it scored 84: Graded for voice cloning, with consent and misuse controls in place of the read-only line and training and retention of voice data in place of the prompt-injection line, since the API returns audio and IDs rather than third-party text. API keys can be limited to chosen endpoints, given a credit quota and an expiry from 15 minutes to 30 days, service-account keys exist, and leaked keys are disabled through GitHub secret scanning (30). PVC is own-voice only and needs a spoken verification matched to the samples, IVC relies on an attestation, high-risk and celebrity voices are blocked, and an AI speech classifier and C2PA support are public (14 of 20). Training on content is on by default with a self-serve opt-out, and the privacy policy caps voice data retention at 3 years after the last interaction (10 of 15). The History API lists each generation with voice, model, characters and date (15). security.txt valid to 2027-03-01 with a disclosure programme, SOC 2 Type II, and a trust centre at compliance.elevenlabs.io, no paid bug bounty found (15 of 20).

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.

## 4. Agent ergonomics, 85 out of 100, up to 2.4 more on the total

Why it scored 85: Creating a clone returns only `voice_id` and `requires_verification`, and `GET /v2/voices` can be narrowed to personal or cloned voices (25). Cursor pagination with `page_size` up to 100 and filters by type, category, fine-tuning state and labels (20). Named error codes with recovery advice for 429, generic statuses elsewhere (15 of 20). No idempotency key, PVC exposes `fine_tuning_state` to poll (10 of 20). Two required fields for an instant clone, and official Python and TypeScript SDKs among others (15).

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.

## 5. Schema & documentation, 87 out of 100, up to 2.1 more on the total

Why it scored 87: Public OpenAPI at api.elevenlabs.io/openapi.json (25). llms.txt and Markdown docs (10). The cloning guides say when to use IVC or PVC and warn that more than 3 minutes of IVC audio can make a clone worse (16 of 20). Inputs are typed, but `labels` is a free-form map (11 of 15). Plenty of examples, while endpoint pages list only 422 and the error codes live in help-centre articles (10 of 15). Versioned paths (`/v1`, `/v2`) and a changelog published several times a month (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. Transparency & trust, 84 out of 100, up to 1.4 more on the total

Made of editorial 75, provenance 92.

Why it scored 84: Closed service with clear terms, SDKs under MIT (15 of 30). Privacy policy updated 20 May 2026 with a 3 year retention cap for voice data, a DPA, subprocessors through the compliance portal and a training opt-out that matches the Terms, but the Terms also take a perpetual, irrevocable licence to User Voice Models while promising deletion on request, which sit uneasily together (25 of 30). Dated deprecations in the changelog, no written notice policy found (15 of 20). Data locations named (United States, the Netherlands, Singapore) and Enterprise residency in the EU, India and Singapore (20).

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)

## 7. Maintenance & community, 85 out of 100, up to 1.3 more on the total

Why it scored 85: Eleven v4 and SDK 2.70.0 on 2026-09-28 (30). Changelog entries roughly weekly and five SDK releases in September (20). Public changelog and a help centre, and SDK issues get labels, but the ten newest open issues show no maintainer replies (10 of 15). Current official SDKs (15). The Python SDK supports 3.8 to 3.15 and shipped three days ago (10).

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.

## 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.

- No bug bounty terms found, only a disclosure programme
- Whether IVC `requires_verification` is ever true for API-created clones, and what triggers it

## Weaknesses

- Instant clones rely on an attestation, not a check
- Training on your content is on by default until switched off in settings
- Terms take a perpetual, irrevocable licence to User Voice Models
- Clones can't be exported
- Only 1 PVC slot until the $299 Scale plan

## 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.

- Keep the source samples. A clone can't be downloaded and re-cloning sounds slightly different
- Check `requires_verification` in the create response before using the voice ID
- List your own clones with `GET /v2/voices?category=cloned` instead of paging the whole library
- On `system_busy` retry with backoff. On `too_many_concurrent_requests` queue instead, the plan cap won't lift
- PVC is asynchronous. Create, add samples, verify, train, then poll `fine_tuning_state`

## What the review panel asked for

- Document `requires_verification` triggers
- Idempotency key on create
- verification on instant clones

## 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.