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

# Fix list: fal image models

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/fal-image, the October 2026 research run, assessed 1 October 2026. Grade B, 65.5 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 image 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 (0). Per-model prices published per image, per megapixel or per token without a login (20). No standing free tier. Free credits are granted case by case (0). Browser sign-up and prepaid credit (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, 65 out of 100, up to 6.1 more on the total

Why it scored 65: Keys carry an API or ADMIN scope, so an agent key can be kept away from deploys and app management. The MCP server also takes OAuth for ChatGPT and web clients. Two coarse scopes, not per-endpoint permissions (25). API scope as the least-privilege mode, no read-only key and no confirmation step (10). Returns generated media and fal's own catalogue text (10). Usage page filters by API key and user (10). No security.txt (0) and no bug bounty found (0). SOC 2 mentioned on the enterprise page, trust centre at trust.fal.ai (10).

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, 73 out of 100, up to 5.4 more on the total

Why it scored 73: Instatus page at status.fal.ai with components and uptime figures, Model API at 99.97 per cent (20). Two incidents in the last 90 days, fal.run unreachable for synchronous requests for 30 minutes on 4 September and 18 minutes of slowness on 29 September. Both under an hour (20). Concurrency published, 2 for new accounts rising to 40 with credit purchases (15). Requests over the limit wait in the queue and server errors aren't charged, but we found no Retry-After or backoff guidance (8). No SLA found (0). GA (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. Agent ergonomics, 72 out of 100, up to 4.6 more on the total

Why it scored 72: The hosted MCP server has 11 tools (15), and search, schema and pricing tools let an agent load one model's definition only when it needs it (5). Outputs come back as CDN URLs, with per-model size and count controls and a header for CDN retention (17). Concurrency errors are typed, but no general error catalogue found (12). No idempotency key, `cancel_job` is the one destructive action, and the MCP docs don't mention readOnlyHint or destructiveHint. Server errors aren't billed, client errors can be if GPU time was spent (8). Official Python and JavaScript clients, few required parameters per model (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. Transparency & trust, 52 out of 100, up to 4.2 more on the total

Made of editorial 32, provenance 71.

Why it scored 52: Closed platform with published terms. The clients are MIT, and each hosted model keeps its own licence (15). The privacy policy (22 July 2026) deletes account data 30 days after closure or after two years idle but says nothing on training with prompts or uploads. The enterprise page says fal never trains on enterprise customers' data, and the FAQ gives CDN retention of at least 7 days (12). No deprecation policy or dated model removals found (0). Processing in the US and other countries with vendor categories named, but no subprocessor list, and the trust centre renders only in a browser (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)

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

Why it scored 85: OpenAPI for the platform APIs at api.fal.ai/v1/openapi.json and a schema per model (25). llms.txt for the docs and a machine-readable llms.txt per model with schema, defaults and price (10). Model pages state what each model does, but quality and when-not-to-use guidance vary by model (14). Typed per-model inputs with enums and defaults (13). Examples on every model page. We found no dedicated errors reference in the docs index (8). Dated changelog, versioned endpoint ids (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, 79 out of 100, up to 1.8 more on the total

Why it scored 79: fal-client 1.0.3 on PyPI on 21 September 2026, 10 days ago (30). Seven dated changelog entries between 8 July and 14 September (20). fal-js has 21 open issues, among them 'fal-js client out of sync with fal api' from January 2026 and a proxy question from November 2025. We couldn't see reply counts (10). Python client current, JavaScript client last released 1.10.1 on 4 May 2026 (12). Recent client releases include supply-chain and retry fixes (7).

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 read the trust centre (browser-only) or reply counts on fal-js issues, so the subprocessor line rests on the privacy policy's vendor categories and responsiveness on open issue ages alone.
- No x402 found, as the listing says.
- fal's official MCP server isn't in the MCP registry under an ai.fal or io.github.fal-ai namespace.

## Weaknesses

- Billing units vary by model, per image, per megapixel or per token
- New accounts limited to 2 concurrent requests
- No security.txt, no bug bounty found and no subprocessor list
- No deprecation policy or dated notices for hosted models
- Client errors can be billed if a runner spent GPU time before failing

## 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.

- Fetch `https://fal.ai/models/<endpoint-id>/llms.txt` before calling a model to get its schema and price
- POST to `https://queue.fal.run/<endpoint-id>` and poll or pass a webhook, rather than holding a call open on fal.run
- Use an API-scoped key for agents. ADMIN keys can deploy and manage apps
- Download outputs you need to keep. CDN retention defaults to at least 7 days, set `X-Fal-Object-Lifecycle-Preference` to change it
- Validate inputs before submitting, a malformed request can still be billed

## What the review panel asked for

- Publish backoff guidance
- Raise new-account concurrency
- state which client errors bill

## 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.