fal music models by fal (Features & Labels, Inc.)

# Fix list: fal music models

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/fal-music, the October 2026 research run, assessed 1 October 2026. Grade B, 68.5 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 fal music models: 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 on fal's endpoints (0). Every model page and fal.ai/pricing show a price and unit without a login, and a pricing API returns unit prices by endpoint ID (20). No free credits or card-free tier documented (0). A person signs up and buys prepaid credit, and the keys API needs an existing ADMIN key (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. Security & auth, 62 out of 100, up to 6.7 more on the total

Why it scored 62: Keys carry one of two scopes, API (call models) or ADMIN (private models, CLI and platform APIs), and are managed in the dashboard and a keys API. Rotation and expiry aren't documented (25). The API scope is a least-privilege key for calling models, and music calls have no destructive actions (10). Returns generated audio, no untrusted content (10). Usage and billing events per request through the platform APIs (10). No security.txt (404 on 2026-10-01), no disclosure policy or bug bounty found. fal's enterprise page mentions SOC 2, and the trust centre at trust.fal.ai renders only in a browser, so we couldn't read the report or scope (7).

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.

## 3. Reliability, 80 out of 100, up to 4 more on the total

Why it scored 80: Instatus status page with components for Model API, Serverless API, Official Models and the dashboards (20). Two incidents in August and September, a 30-minute outage of synchronous calls to fal.run on 2026-09-04 (queue calls weren't named) and an 18-minute slowdown on 2026-09-29. Neither reached an hour. July wasn't shown in the history (20). Concurrency limits published, 2 for a new account rising to 40 self-serve with spend (15). A 429 carries type `concurrent_requests_limit` and an `X-Fal-needs-retry` header, the clients retry with exponential backoff, and queued requests are never rejected for concurrency (8). Failed requests and queue time aren't charged, which makes retries safe to spend on (7). No SLA published for self-serve (0). The queue 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.

## 4. Transparency & trust, 59 out of 100, up to 3.6 more on the total

Made of editorial 47, provenance 71.

Why it scored 59: Closed platform with terms. Output rights come from each upstream model's licence, and model pages mark commercial use, but we couldn't read the terms on 2026-10-01 to confirm how they pass through (12). Privacy policy updated 2026-07-22 deletes data 30 days after an account closes and after 2 years inactive, but doesn't say whether prompts train models, and we found no DPA (15). Dated endpoint deprecations, such as `fal-ai/elevenlabs/music` on 2026-12-17 and the legacy CDN cleanup on 2026-06-01, with no written policy (15). Data is processed in the US and other countries, and no subprocessor list was found (5).

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: fal.ai, registered 2020-11-13 (5 years) (11 of 15)
- Endpoint on the vendor's domain: queue.fal.run is not on fal.ai (0 of 15)
- security.txt: not found (0 of 10)

## 5. Agent ergonomics, 80 out of 100, up to 3.3 more on the total

Why it scored 80: Results are small JSON with a file URL (15). Length and format controls depend on the model, Lyria 2 is fixed at 30 seconds while MiniMax and ElevenLabs take a duration (5). Queue status, webhooks and a catalogue API with `limit` and search (20). Typed 429s and schema validation errors (15). Failed requests and queue time aren't charged and the clients retry, but there's no idempotency key (10). Most music models need only a prompt (7), official JavaScript and Python clients (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.

## 6. Schema & documentation, 85 out of 100, up to 2.4 more on the total

Why it scored 85: Per-endpoint OpenAPI documents for the queue API, for example `fal-ai/lyria2` (25). llms.txt for the docs and a per-model llms.txt with input and output schemas (10). Model pages say what each model is for, but rarely when not to use it, and quality varies by model (12). Inputs are typed per model, with enums such as `output_format` and ranges such as `music_length_ms` from 3,000 to 600,000 on ElevenLabs v2.5 (13). Code examples on every model page, the concurrency 429 documented, other errors less so (10). Versioned endpoint IDs (`elevenlabs/music/v2.5`) and a dated public 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.

## 7. Maintenance & community, 82 out of 100, up to 1.6 more on the total

Why it scored 82: fal-client 1.0.3 on PyPI on 2026-09-21, changelog entry on 2026-09-14 (30). Seven dated changelog entries between 2026-07-08 and 2026-09-14 (20). Public changelog and dashboard support, response times not checked (10). Python client current at 1.0.3, JavaScript client last released as 1.10.1 on 2026-05-04 (12). The Python package publishes from GitHub Actions with attestations (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.

- We couldn't load the terms of service on 2026-10-01, so how upstream model licences pass through to fal customers is taken from model pages, not the terms.
- July 2026 wasn't shown in the status history.
- We didn't re-verify every per-model price in the listing's pricing notes against today's model pages, only Lyria 2 and Beatoven.
- The SOC 2 report behind trust.fal.ai couldn't be read, so its scope and date are unknown.

## Weaknesses

- Billing units differ per model (per track, per second, per minute rounded up)
- Output licence and content rules vary by upstream model
- No free tier and no security.txt or disclosure policy
- Unversioned endpoints get deprecated, `fal-ai/elevenlabs/music` on 2026-12-17

## 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.

- Submit to `queue.fal.run` and poll or use a webhook. Synchronous `fal.run` calls went down for 30 minutes on 2026-09-04
- Check the unit on the model page. ElevenLabs v2.5 bills a whole minute for a 30-second clip
- Pin the versioned endpoint, for example `fal-ai/elevenlabs/music/v2.5`
- Use an API-scope key for agents, never an ADMIN key
- For stems, run `fal-ai/demucs` on the output

## What the review panel asked for

- Idempotency key on submit
- Self-serve first key
- Normalise price units

## 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.