Fish Audio Voice Cloning API by Fish Audio

# Fix list: Fish Audio Voice Cloning API

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/fish-audio-voice-cloning, the October 2026 research run, assessed 1 October 2026. Grade D, 51.5 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 Fish Audio Voice Cloning 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. Security & auth, 28 out of 100, up to 12.6 more on the total

Why it scored 28: 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. Plain revocable API keys with no scopes found (20 of 30). No consent field or speaker check in the API, only guidance to clone your own voice or one you have written permission for, and no watermark or detection tool found (0 of 20). The terms take a perpetual, irrevocable licence to submissions including for model training, with no opt-out, and the privacy policy keeps content as long as needed (0 of 15). Credit and wallet usage readable through the API, no per-call log found (8 of 15). No security.txt, disclosure policy, bug bounty, SOC 2 or trust centre found (0 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.

## 2. 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 are public, $15 per million UTF-8 bytes of speech and $0.01 per successful voice design request, with no published fee to create a voice model (20). The `s2.1-pro-free` model costs $0 under fair use, the card requirement isn't stated (10 of 20). Signup is a human browser flow and the API is prepaid (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.

## 3. Maintenance & community, 13 out of 100, up to 7.6 more on the total

Why it scored 13: The newest dated release we found is SDK 1.3.0 on 2026-03-10 and the newest changelog entry is March 2026, so nothing dated in the last 180 days (0). No dated releases or entries in the last 90 days (0). A changelog exists but isn't kept up (5 of 15). Official SDKs in Python and JavaScript, the Python one last released 205 days ago (8 of 15). SDK older than 90 days (0 of 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.

## 4. Reliability, 65 out of 100, up to 7 more on the total

Why it scored 65: Status page at status.fish.audio (Better Stack) with Platform API, TTS API and per-model components and 90 days of uptime (20). One incident in the last 90 days, 10 minutes of TTS API downtime from an APAC data centre on 10 August (20). Concurrency limits are published by prepaid spend, 5, 15 and 50 concurrent requests (15). No 429, Retry-After or backoff guidance found (0). The models page mentions latency guarantees for s2.1-pro, but no SLA document was found (0). Cloning, s2.1-pro 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.

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

Why it scored 65: Creating a model returns the model object with author and engagement fields, heavier than an ID and status (20 of 25). A list endpoint exists, we didn't confirm its pagination or an own-voices filter (10 of 20). Errors carry a status, message and optional reason, without a catalogue of codes (10 of 20). No idempotency key, but inline cloning needs no create step and voice design bills only successful requests, so a failed call can be retried at no cost (10 of 20). Official Python and JavaScript SDKs and a short required field list (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.

## 6. Schema & documentation, 79 out of 100, up to 3.4 more on the total

Why it scored 79: Public OpenAPI at docs.fish.audio/api-reference/openapi.json (25). llms.txt and Markdown pages (10). The docs explain when to create a persistent model and when to pass reference audio inline (14 of 20). Inputs typed, with enums for `visibility` and `train_mode`, 1 to 20 voice files, and ranges on every voice design field (14 of 15). Examples throughout, but the create-model reference documents only 401 and 503 with a `{status, message, reason}` body (8 of 15). There's a changelog, but its newest entry is March 2026 and it misses s2.1-pro and voice design, and paths mix unversioned `/model` with `/v1` (8 of 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.

## 7. Transparency & trust, 61 out of 100, up to 3.4 more on the total

Made of editorial 40, provenance 82.

Why it scored 61: Closed API with clear terms, plus S2-Pro weights published under Fish Audio's own research licence, which isn't OSI-approved (20 of 30). The terms (effective 18 August 2024) take a perpetual, irrevocable licence and the privacy policy publishes no retention period, and we found no DPA or subprocessor list (5 of 30). A deprecations page with dates, Fish Speech v1.5 and v1.6 retired on 2026-02-28 (15 of 20). No subprocessors or data locations found (0 of 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: fish.audio, registered 2023-12-11 (2 years) (7 of 15)
- security.txt: not found (0 of 10)

## 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.

- Whether `GET /model` paginates and filters to your own models, we didn't confirm the parameters
- The release date of s2.1-pro and voice-design-1, neither is in the changelog
- Whether the free model needs a card or a prepaid balance
- security.txt and certifications were taken as absent from last week's check and today's pages, we didn't re-fetch security.txt

## Weaknesses

- No consent or speaker verification in the API
- Terms take a perpetual, irrevocable licence to uploads, including for training, with no opt-out
- Changelog stops at March 2026, s2.1-pro and voice design aren't in it
- No 429 or retry guidance despite concurrency caps of 5 to 50
- No security.txt, SOC 2 or subprocessor list found

## 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.

- Send 2 or 3 clean single-speaker clips with matching `texts`, or the API runs ASR on them
- Pass the model `_id` as `reference_id` in TTS. Check `state` before relying on it
- For one-off voices send `references` inline to `/v1/tts` instead of creating a model
- Send the `model: voice-design-1` header on voice design calls
- Keep concurrency at 5 until prepaid spend passes $100, there's no documented retry guidance

## What the review panel asked for

- 429 handling guidance
- Current changelog
- consent verification
- a training opt-out

## 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.