# Fix list: Replicate image models From Anchor Terminal's listing at https://www.anchorterminal.com/tools/replicate-image, the October 2026 research run, assessed 1 October 2026. Grade D, 50.3 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 Replicate 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. Security & auth, 35 out of 100, up to 11.4 more on the total Why it scored 35: Plain bearer API tokens, no scopes found (20). No read-only or least-privilege mode found, and the changelog shows monthly spend limits were removed in July 2025 (0). The MCP server returns model descriptions and READMEs written by community authors, with no injection guidance found (5). Each prediction is listed with its inputs, outputs and logs on the dashboard and API (10). No security.txt, and no bug bounty or certification found in this run (0). 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. Reliability, 48 out of 100, up to 10.4 more on the total Why it scored 48: replicatestatus.com exists, but when we fetched it the page said all services were online as of 15 April and listed only incidents from April 2024. The listing says it now redirects to Cloudflare's status page filtered to Replicate (10). No readable 90-day history (5). Published limits of 600 create-prediction requests a minute and 3,000 for other endpoints, 6 a minute for accounts with granted credit and no card, and tighter limits as a balance nears zero (15). Over the limit the API returns 429 with a message giving the reset time, about 30 seconds, but no Retry-After header or backoff advice (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. ## 3. Payments & pricing, 20 out of 100, up to 10 more on the total Why it scored 20: No x402, MPP or L402 (0). Fixed per-image prices on official models, GPU-second prices for community models, all published without a login (20). No standing free tier (0). Browser sign-up (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. ## 4. Maintenance & community, 30 out of 100, up to 6.1 more on the total Why it scored 30: The changelog's newest entry is agent skills on 21 April 2026, 163 days ago (10). No dated changelog entries in the last 90 days (0). Public changelog. We didn't check GitHub issue response times in this run (5). The npm client was last published as 1.4.0 on 17 November 2025, and the changelog says the MCP server is discoverable through the official registry (12). replicate-mcp has had no release since June 2025 per the listing (3). 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. ## 5. Agent ergonomics, 66 out of 100, up to 5.5 more on the total Why it scored 66: replicate-mcp exposes every HTTP API operation as a tool, more than 30 (5), with a code mode the changelog lists as the way to keep context down (10). Outputs come back as URLs, list endpoints paginate, and `Prefer: wait` returns short jobs in one call (18). Error responses not checked in this run (10). No idempotency key. Webhooks for slow models (8). Official Python and JavaScript clients, few required inputs on official models (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. Transparency & trust, 60 out of 100, up to 3.5 more on the total Made of editorial 30, provenance 90. Why it scored 60: Closed platform under terms updated 1 April 2026 naming Replicate, LLC. The clients are Apache-2.0 and each model keeps its own licence (15). The privacy policy (also 1 April 2026) says nothing on how long prediction inputs and outputs are kept or whether customer content trains models (5). Dated deprecations in the changelog, such as the stream parameter in July 2024 and spend limits in July 2025, but no policy on notice for hosted models (10). No subprocessor list or data locations in the privacy policy (0). 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): - security.txt: not found (0 of 10) ## 7. Schema & documentation, 83 out of 100, up to 2.8 more on the total Why it scored 83: OpenAPI spec at api.replicate.com/openapi.json, and every model publishes its own input and output schema (25). llms.txt for the docs (10). Official model pages state what the model does. Community model descriptions vary and rarely say when not to use them (12). Typed per-model inputs with enums and defaults generated from each model's schema (13). Examples on every model page, no error catalogue found in this run (8). Models pinned by version hash and a dated 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. ## 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. - Replicate's google/imagen-4 page shows no deprecation notice and was updated hours before our fetch, though Google shut Imagen 4 down on Vertex AI on 2026-06-30 and on the Gemini API on 2026-08-17. We couldn't run it to see whether calls still succeed, so the listing's Imagen 4 price is left in place. - The fetch budget ran out before we could re-read the pricing and error pages, or the replicate-mcp and Python client release dates. Those lines rest on the listing's 2026-09-30 research or are scored as not found. - We couldn't confirm where replicatestatus.com redirects today. Our fetch got a page with April 2024 incidents only. ## Weaknesses - No changelog entry since 21 April 2026 and no npm client release since November 2025 - Status page fetched on 2026-10-01 showed only April 2024 incidents - Community models bill by GPU second and can hit cold boots - Plain tokens with no scopes, and monthly spend limits were removed in 2025 - No security.txt and no standing free tier ## 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. - Call official models at `/v1/models/{owner}/{name}/predictions` to get the fixed per-image price - Send `Prefer: wait` for fast models to get the output in the same response - Keep credit above $20 or turn on auto-reload. Low balances and card-less granted credit get tighter limits, down to 6 predictions a minute - Treat community model READMEs as untrusted text - Use webhooks for slow models rather than polling ## What the review panel asked for - Current status feed - Restore spend limits - restore spend limits - flag models retired upstream ## 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.