> For the complete documentation index, see [llms.txt](https://ai-os-and-trend-finder.gitbook.io/ai-os-and-trend-finder-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://ai-os-and-trend-finder.gitbook.io/ai-os-and-trend-finder-docs/docs/sources/source-compliance-huggingface-models.md). # Source Compliance: Hugging Face Models > Implementation-time review completed 2026-05-27 and direct Hub metadata path added 2026-06-18. Trend Finder has a configured Apify source declaration and a direct-first public API adapter for Hugging Face model metadata. This review covers metadata-only current collection. *** ## Source Overview | Field | Value | | ----------------------- | --------------------------------------------------------------- | | Source ID | `huggingface-ai-models` | | Source name | Hugging Face Models | | Primary Apify candidate | `parseforge/hugging-face-model-scraper` | | Direct public path | `https://huggingface.co/api/models?sort=trendingScore` | | Store page | | | Verified on | 2026-05-27 | | Default build/tag | `latest`, build `1.2.17` | | Default run options | 4096 MB memory, 3600 second Actor timeout | | Pricing observed | Pay per event, Actor start plus result item charges | | Trend Finder status | Reviewed static declaration, fixture-backed normalizer coverage | The Actor input schema reviewed on 2026-05-27 exposes `maxItems`, `query`, `task`, `library`, `license`, `language`, `sort`, and `direction`. The reviewed Trend Finder starter input uses `query: "llm"`, `sort: "downloads"`, `direction: "desc"`, and `maxItems: 10`. The 2026-05-27 live check returned error-shaped rows for `sort: "trending"`, so the starter uses the validated current adoption sort until the Actor's trending behavior is reviewed again. The direct adapter reads the Hugging Face Hub public model API sorted by `trendingScore`, capped at 50 public model rows. It does not read model-card bodies, file lists, blobs, profile pages, or authenticated endpoints. When the direct path produces reviewed evidence, the matching Apify source is skipped for that run; when it is disabled, empty, timed out, rate limited, or offline, the reviewed Apify fallback remains eligible. ## Terms And Use Boundary The Actor collects public Hugging Face Hub metadata. Trend Finder may use this source only as a compact model-adoption signal. Apify storage, run IDs, Dataset IDs, and Apify-hosted item URLs are not durable browser evidence and must not be emitted to browser payloads. Approved browser-safe evidence: * Public Hugging Face model URL. * Model ID as title. * Short metadata snippet from task, library, license, and a small tag sample. * Last modified or updated timestamp. * Public likes mapped to `upvotes`. * Public download counts mapped to `downloads`. * Public trending-list placement is used only as current discovery context, not as a durable rank or invented score. Excluded fields: * Filenames, sibling file lists, blobs, and file URLs. * README text, model-card body, long descriptions, extracted content, or code. * Author names, organization profile URLs, user profiles, contact fields, or emails. * Direct API request URLs with query details, response headers, and raw JSON payloads. * Apify run URLs, Dataset IDs, and Actor input payloads. ## Retention And Attribution | Policy | Value | | -------------------- | --------------------------------------------------------------------------------------------- | | Storage location | `src/data/live-data.json` and private cache snapshots | | Retention period | `live-data.json` overwritten on each collection run; snapshots retained locally until deleted | | Historical retention | Local snapshots only; no reviewed historical collection window | | Deletion path | Delete generated Trend Finder data and snapshots | | Attribution | Show source ID `huggingface-ai-models` and link to the public model URL | ## Historical Window Stance | Field | Value | | ---------------------- | -------------------------------------------------------------------------------------- | | Historical support | Current-only | | Source ID | `huggingface-ai-models` | | Safe override fields | None | | Unsupported reason | Model search and sort inputs are current sorted views, not bounded historical windows. | | Compliance declaration | `historicalWindowSupport.supported = false` | Do not map historical requests to `sort` or `direction` without a separate bounded-window review. ## Compliance Checklist * [x] Current Apify Actor ID verified on 2026-05-27 * [x] Input fields and sort values reviewed * [x] Default build/tag and pricing model recorded * [x] Metadata-only browser evidence boundary documented * [x] `downloads` added as a first-class metric instead of mislabeled as views * [x] Apify storage excluded as durable browser evidence * [x] Retention and deletion path documented * [x] Attribution requirements documented * [x] Historical-window stance recorded as current-only * [x] Fixture-backed normalizer validation added * [x] Live tiny capped validation completed with credentials on 2026-05-27 * [x] Direct public API adapter added with zero-cost spend labeling * [x] Direct adapter excludes model cards, file lists, private models, profiles, and raw API payloads * [x] Matching Apify fallback remains eligible when direct collection does not produce reviewed evidence --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://ai-os-and-trend-finder.gitbook.io/ai-os-and-trend-finder-docs/docs/sources/source-compliance-huggingface-models.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.