# Fix list: Black Forest Labs FLUX API From Anchor Terminal's listing at https://www.anchorterminal.com/tools/black-forest-labs, the October 2026 research run, assessed 1 October 2026. Grade B, 64.2 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 Black Forest Labs FLUX 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. Payments & pricing, 20 out of 100, up to 10 more on the total Why it scored 20: No x402, MPP or L402 (0). Per-image and per-megapixel prices published in the docs without a login (20). No free credits documented (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. Reliability, 65 out of 100, up to 7 more on the total Why it scored 65: Atlassian Statuspage at status.bfl.ai (redirects to status.bfl.ml) with a component per model family and per regional endpoint (20). The last 90 days hold two majors and several long slowdowns. The FLUX.2 [flex] cluster was down for about 7.5 hours on 19 August and an EU cluster outage ran about 4.75 hours on 20 August with traffic shifted elsewhere. US latency ran 6.5 hours on 8 September and EU latency about 20 hours on 29 and 30 September. Two majors sit between the rubric's 10 for one and 0 for several (5). 24 concurrent requests on most endpoints, 6 on flux-kontext-max (15). Backoff on 429 documented with example code (15). An SLA is mentioned for enterprise contracts with no figure published (0). FLUX.2 is 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. Security & auth, 65 out of 100, up to 6.1 more on the total Why it scored 65: API keys scoped to projects, several per project, shown once. The MCP server uses OAuth with organisation selection. No permission scopes on keys (25). Projects separate usage, but we found no per-project spend cap or read-only key (5). Output is generated images only (10). Organisation activity logs and per-project usage metrics, no per-request log (10). security.txt valid (5), SOC 2 Type II and ISO 27001 claimed with a Vanta trust centre at trust.bfl.ai (10), no bug bounty found (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. ## 4. Agent ergonomics, 67 out of 100, up to 5.4 more on the total Why it scored 67: Responses carry a short-lived URL, not image bytes, and the MCP server has 7 tools (22). Width, height and output format controls, history pagination in MCP (15). Status codes and task statuses an agent can act on, with moderation reasons added in June 2026 (17). No idempotency key, so a resubmitted job is a new paid job, and the MCP docs don't mention readOnlyHint or destructiveHint (5). Only `prompt` required on generation, but no official SDK package (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. ## 5. Transparency & trust, 66 out of 100, up to 3 more on the total Made of editorial 40, provenance 92. Why it scored 66: Closed API with clear developer terms that give you the output (15). The privacy policy (1 August 2026) says prompts and uploads may be used for training unless you email an opt-out, while the enterprise page promises zero data retention on the managed API. No retention periods stated (10). No deprecation policy, and the developer terms (4 August 2026) allow changing or removing parts of the service without notice (5). EU and US regional endpoints and named controllers per region, but no subprocessor list (10). 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: bfl.ai, registered 2022-06-29 (4 years) (7 of 15) ## 6. Maintenance & community, 73 out of 100, up to 2.4 more on the total Why it scored 73: Last release note on 10 September 2026, 21 days ago (30). For models we look at churn and notice in place of release counts. FLUX.1 models stayed on sale after FLUX.2 shipped, but there's no notice policy and the terms allow changing or removing parts of the service without notice (15). Public release notes, support channel not tested (8). Official MCP server in the registry as io.github.black-forest-labs/flux-mcp, published 26 May 2026 (15). No SDK packages to assess (5). 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. Schema & documentation, 88 out of 100, up to 2 more on the total Why it scored 88: OpenAPI specs at api.bfl.ai/openapi.json plus two more for licences and video edit (25). llms.txt with about 120 links, every page also served as Markdown (10). Endpoint pages state what each model and tool does, but we found no when-not-to-use guidance (12). Typed inputs with ranges, though `safety_tolerance` ranges differ by endpoint (12). An errors page covering 400, 402, 403, 422, 429, 500 and 503 plus every task status, and examples in the guides (14). Versioned path and a dated release notes page (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. - No published figure for the enterprise SLA. - We couldn't establish whether a request that ends as Request Moderated or Content Moderated is billed. - The listing's lastRelease of 2026-05-26 was the MCP registry date. Release notes run to 2026-09-10, corrected in patch. ## Weaknesses - Two cluster outages over four hours (19 and 20 August 2026) and slowdowns of 6.5 and 20 hours since - Result URLs expire after 10 minutes - Prompts and uploads may be used for training unless you email an opt-out - Developer terms allow changing or removing parts of the service without notice - No official SDK package and no free credits ## 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. - Poll the `polling_url` from the submit response, never a hand-built URL - Download `result.sample` straight away. It's gone after 10 minutes - Stay under 24 concurrent jobs, 6 on flux-kontext-max, and back off exponentially on 429 - Treat 402 as out of credit, not as a retryable error - Use api.eu.bfl.ai or api.us.bfl.ai only when data must stay in that region, the global endpoint fails over ## What the review panel asked for - Idempotency key on submit - Longer result retention - state whether moderated requests 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.