LibreTranslate by LibreTranslate

# Fix list: LibreTranslate

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/libretranslate, the October 2026 research run, assessed 1 October 2026. Grade D, 49.8 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 LibreTranslate: 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. Reliability, 35 out of 100, up to 13 more on the total

Why it scored 35: Read as the hosted API at libretranslate.com, since the listing is an http-api. No status page found (0). No readable incident history (5). Plan limits published with numbers, 80 calls a minute burst and about 20 sustained on Pro, 200 and about 50 on Business, 2,000 characters a call (15). The spec documents 429 with a Slowdown message, and repeated limit violations get a 403 ban, but there's no Retry-After or backoff advice (5 of 15). No SLA (0). Release 1.9.6 is a stable release, not beta (10). Self-hosted, the package reading would score far higher, with CI across Python 3.8 to 3.14 and a 1.x release line.

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.

## 2. Security & auth, 45 out of 100, up to 9.6 more on the total

Why it scored 45: Model reading of the checklist, with training and retention in place of the least-privilege and injection lines. One API key per hosted plan, sent in the request body, with no scopes or rotation found. A self-hosted server can run keyless or require keys (10 of 30). The hosted privacy policy says translation texts aren't stored or logged, so nothing is kept to train on (20). IP addresses are kept for 2 days against abuse, no text retained (15). No usage log or per-call view found for the hosted plans (0). No security.txt, SECURITY.md, disclosure policy or bug bounty found (0). Self-hosting under AGPL-3.0 keeps all text on your own network.

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. Payments & pricing, 40 out of 100, up to 7.5 more on the total

Why it scored 40: No x402, MPP or L402 (0). Flat monthly plans with published limits, $29 Pro and $58 Business, not per-unit pricing (10). No hosted free tier or trial, only a 7-day money-back guarantee. We depart from the rubric and give the free line for the self-hosted server, which runs the same API free and with no card (20). An agent can `pip install libretranslate` and run a keyless server, but the hosted plans need a person at a Stripe checkout (10 of 20).

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.

## 4. Schema & documentation, 65 out of 100, up to 5.7 more on the total

Why it scored 65: Swagger 2.0 document at libretranslate.com/spec with Swagger UI at /docs, versioned with the release (25). No llms.txt at docs.libretranslate.com (0). Endpoint descriptions are one line each and don't say when to use detect, translate or translate_file (10 of 20). `format` is an enum (text, html) and required fields are marked, but `q` takes a string or an array without a typed union (10 of 15). Each operation lists 200, 400, 403, 429 and 500 with meanings. We didn't check the docs site for worked examples (10 of 15). GitHub releases and tags, no CHANGELOG file (10 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.

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

Why it scored 65: API reading of the checklist. Responses are the translated text, the detected language when `source` is auto, and `alternatives` only when asked (20 of 25). `q` takes an array for several strings, but 2,000 characters a call on the hosted plans and no other output controls (12 of 20). Errors are JSON with a readable message, such as a language that isn't supported or a missing parameter, but no codes (15 of 20). Translation is stateless, but the hosted service bans a client after repeated limit violations and gives no retry timing (10 of 20). Three parameters, `source` can be auto, but no official client SDK confirmed beyond the server package (8 of 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. Maintenance & community, 40 out of 100, up to 5.3 more on the total

Why it scored 40: Last release v1.9.6 on 2026-05-26, 128 days ago (10). No release in the last 90 days, though 18 commits landed since 3 July, the latest on 27 September moving translated text out of the URL query string (0). 127 open issues. The newest ones from July to September carry labels from an automated triage bot and some have 1 to 4 comments, and a contributor's fix was merged on 27 September (15 of 25). Current PyPI and Docker packages, no official client SDK confirmed (5 of 15). CI runs pytest on Python 3.8 to 3.14 plus Docker builds and health checks (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.

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

Made of editorial 55, provenance 66.

Why it scored 61: Open source under AGPL-3.0 (30). The privacy policy, updated 23 September 2025, says no text is stored or logged, IPs are kept 2 days and no tracking is used, but there are no terms of service or DPA for the hosted API (15 of 30). No deprecation policy or dated notices found (0). We found no telemetry in the server code. For the hosted service Stripe is named, but not the hosting provider or data location (10 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: libretranslate.com, registered 2020-12-20 (5 years) (11 of 15)
- Terms of service: not found (0 of 10)
- Status page: not found (0 of 10)
- 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.

- Reliability is read for the hosted service. A self-hosted deployment's reliability is the operator's own.
- unchecked: whether hosted API keys can be rotated or revoked from the portal.
- We couldn't count maintainer replies across the 127 open issues from the issues page.

## Weaknesses

- No status page, SLA or terms of service for the hosted API
- 2,000 characters a call and about 20 calls a minute sustained on Pro
- No glossaries or formality control, about 50 languages
- No release since 1.9.6 on 2026-05-26
- No security.txt or disclosure policy

## 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.

- Split text into calls of 2,000 characters or fewer on the hosted service
- Pace to about 20 calls a minute on Pro, since repeated 429s turn into a 403 ban
- Pass `source=auto` to detect and translate in one call
- Set `format=html` for markup
- Read /spec on the server you call, since self-hosted limits and options differ from the hosted ones

## What the review panel asked for

- Add a hosted trial key
- glossary support
- error codes

## 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.