# Fix list: Glimpse From Anchor Terminal's listing at https://www.anchorterminal.com/tools/elorian-glimpse, the October 2026 research run, assessed 17 October 2026. Grade D, 48.6 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 Glimpse: 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, 38 out of 100, up to 12.4 more on the total Why it scored 38: Hosted lines, read on the OpenRouter route. OpenRouter has a status page at status.openrouter.ai, which covers the platform rather than this model, and Elorian has none found (10 of 20). The endpoint is a day old. OpenRouter's model API reports 98.7 per cent uptime over the last day and 100 per cent over the last 30 minutes on 17 October 2026, which is not a 90-day record (5). OpenRouter's limits page documents credit, budget and rate limits and a key endpoint that reports what's left, but sets no request cap on paid models, and no limit for Glimpse or the Together deployment was found (8 of 15). The errors page documents Retry-After on 429 and 503 and on 402 in-flight budget errors, and says the official SDKs honour it (15). No SLA from Elorian or for this endpoint (0). Elorian labels Glimpse a research preview (0). 38. 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. Payments & pricing, 25 out of 100, up to 9.4 more on the total Why it scored 25: No machine payment protocol on OpenRouter or from Elorian (0). Per-token prices published without a login on OpenRouter and in Elorian's launch post (20). Glimpse has no free variant on OpenRouter. Elorian's post says it can be tried free on Elorian's platform, which needs a sign-in, and whether a card is needed was not established (5 of 20). A person signs up in a browser on either route, and OpenRouter's programmatic keys come only after that (0). 25. 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. ## 3. Security & auth, 56 out of 100, up to 7.7 more on the total Why it scored 56: Model reading. OpenRouter bearer keys with per-key credit limits, management keys and OAuth PKCE, no permission scopes (25 of 30). Training on API data on this route is off according to OpenRouter's policy display for Together (no training, no prompt retention). Elorian's terms licence all User Content for training, with an email opt-out for paid users, and its privacy policy trains on free use by default. Whether Elorian receives prompts from the Together deployment was not established (8 of 20). OpenRouter's zero-retention filter and Together's no-retention flag, against Elorian's retention 'for as long as reasonably necessary' with no period (8 of 15). OpenRouter's per-generation activity records for the operator (10 of 15). OpenRouter's security.txt was valid at its last check. Elorian has no security.txt, disclosure policy, bug bounty or certification found (5 of 20). 56. 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. Schema & documentation, 60 out of 100, up to 6.5 more on the total Why it scored 60: Model reading. Elorian publishes no API reference, model card or llms.txt in public. Its console docs need a sign-in. The contract an agent has is OpenRouter's generic OpenAPI-documented chat endpoint plus the model API's list of supported parameters, which is not specific to Glimpse (20 of 25). OpenRouter's llms.txt and Markdown docs, including video input (10). The launch post and OpenRouter's description name the jobs (visual search, video captioning, dense counting, robotics planning) and the post states limits (token cost, no action outputs yet), but nothing says when not to use it or how video is sampled (6 of 20). Typed parameters on OpenRouter, with `tool_choice` limited to none, auto and required, and no documented reasoning effort levels for this model (10 of 15). OpenRouter's examples and error reference, with no Glimpse-specific examples or limits on video length or image count (7 of 15). OpenRouter's dated changelog and a dated canonical slug, `glimpse-20261016`, but no changelog from Elorian (7 of 15). 60. 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. Transparency & trust, 36 out of 100, up to 5.6 more on the total Made of editorial 31, provenance 40. Why it scored 36: Closed model under clear terms (California law, arbitration with an opt-out, liability caps), with weights not released and the open-weight base model unnamed (13 of 30). The privacy policy and terms agree that Elorian trains on submitted content, but disagree on who may opt out (anyone by privacy@elorian.ai in the policy, paid users by contact@elorian.ai in the terms), say only that 'different terms may apply' to API use, and give no retention periods. The published privacy and acceptable use pages carry a leftover drafting comment from a terms outline, and the acceptable use page is titled Privacy Policy (10 of 30). No deprecation policy or dated notices, and the terms allow discontinuation without notice (0 of 20). Services hosted in the United States and subprocessors named by category only. On this route, OpenRouter names Together as the provider and shows its data policy (8 of 20). 31. 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: elorian.ai, registered 2025-12-03 (under a year) (0 of 15) - Endpoint on the vendor's domain: openrouter.ai is not on elorian.ai (0 of 15) - Status page: not found (0 of 10) - Changelog: not found (0 of 10) - security.txt: not found (0 of 10) ## 6. Agent ergonomics, 66 out of 100, up to 5.5 more on the total Why it scored 66: Model reading of the checklist (tool use, structured output, caching, context, batch, SDKs, errors). Function tools with `tool_choice` none, auto or required, but not a named function, and the multimodal tools and subagents the launch post describes are not exposed (`native_tools` is empty) (12 of 20). `response_format` and `structured_outputs` are listed, with enforcement depending on Together's deployment and no Elorian documentation (10 of 15). Implicit prompt caching at $0.06 per million cached input tokens, a tenth of the input price, with OpenRouter's sticky routing (12 of 15). 262,144-token context with up to 256,000 output tokens (12 of 15). No batch endpoint for this model, since its only endpoint has no `:batch` variant (0 of 10). OpenRouter's SDKs in TypeScript, Python and Go, and any OpenAI client works (10 of 10). OpenRouter's documented error codes with `limit_source` and remedy hints, nothing specific to video or image limits (10 of 15). 66. 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. ## 7. Maintenance & community, 51 out of 100, up to 4.3 more on the total Why it scored 51: Model reading. Announced on 16 October 2026 and listed on OpenRouter on 17 October (30). One release, no history of versions or retirements, no deprecation notice period, and Elorian's terms let it modify or discontinue the Service without notice (3 of 20). Elorian's support page says there is no support queue, the terms say there is no obligation to support, and channels are email and a Discord server, not tested (6 of 25). No official SDK from Elorian. OpenRouter's SDKs are current (8 of 15). No package or CI of Elorian's to read. OpenRouter's SDK repositories run CI (4 of 10). 51. 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. - unchecked: Elorian's own API documentation, rate limits and per-key controls, which sit inside the signed-in console at platform.elorian.ai/settings/docs. The page script that carries them was not used as evidence - unchecked: status.elorian.ai, which our shell could not reach, so whether Elorian has a status page was not established - unchecked: whether Elorian receives or can train on prompts sent through OpenRouter to the Together deployment. OpenRouter shows Together's policy as no training and no retention, and Elorian's privacy policy says different terms may apply to paid and API use without naming them - unchecked: whether Elorian's free trial needs a card, and how large it is - unchecked: which open-weight base model Glimpse was post-trained on. The launch post doesn't say, so the base model's licence terms are unknown - Elorian's terms forbid using the Service through any agent or automated software other than Elorian's own, and the acceptable use policy forbids publishing benchmark or performance information without consent. Both are recorded as facts. They apply to Elorian's own Service, and how they apply to OpenRouter's route was not established - The launch post's claims of frontier-level performance 'at the level of (and sometimes exceeding)' GPT-6 Astra and Claude Opus 5.5 rest on Elorian's own table, which shows Glimpse above Astra on four of twelve evaluations. The price comparisons match the prices in our openai-api, anthropic-api and gemini-api listings. No deduction taken - The scout's entry (research/scout-2026-10-17/models.md 9) was right on the facts checked. Its OpenRouter listing date of 17 October matches OpenRouter's release date of 17 October 2026 03:10 UTC ## Weaknesses - Labelled a research preview, with no public API docs, model card, changelog or deprecation policy from Elorian - One provider on OpenRouter (Together, on a dedicated deployment), so there is no fallback if that endpoint goes down - Elorian's terms licence submitted content for training. Paid users opt out by email, and free use trains by default - Elorian's terms forbid using the Service through any agent or automated software other than Elorian's own - The tools the launch post says Glimpse is trained to use (code, image transforms, video watching, subagents) aren't exposed on OpenRouter ## 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 it through OpenRouter as `elorian/glimpse` at https://openrouter.ai/api/v1/chat/completions. Elorian's own API docs sit behind a console sign-in - Send video as a `video_url` content part, as a public URL or a base64 data URL. Batch requests are not available for this model - Set `max_tokens` yourself. The endpoint allows up to 256,000 output tokens, and visual reasoning spends many of them - Honour Retry-After on 429 and 503. With a single provider, a 503 means the Together deployment is unavailable, not that another host can take over - Treat counts and timestamps as unverified. Elorian's terms say outputs need independent human review before anyone relies on them ## What the review panel asked for - Publish typical reasoning lengths - Publish API docs and rate limits - Publish an API reference - Document video sampling and limits - Publish model rate limits - Add a second provider - state API data terms - publish a disclosure policy ## 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.