Web URL & HTML Distillation with BM25 Budget Control

Choose this endpoint when you need to feed web content to an LLM and want to minimize token consumption with measurable ROI metrics. It is especially valuable when you have a strict token budget (use mode=budget with a query and max_tokens), need structured table/metadata extraction from HTML (mode=structured), or want a fast outline of page structure (mode=skeleton). Prefer it over generic scrapers when you need built-in BM25 relevance filtering, compression ratio reporting, and LLM cost savings quantification. It also accepts raw HTML directly, making it useful when you already have scraped content and just need distillation without an extra fetch.

• URL unreachable or returns non-200 status — endpoint cannot fetch content
• HTML payload too large for processing within token constraints
• Invalid mode enum value provided — falls back to markdown or returns error
• max_tokens set below minimum of 20 — validation error returned
• BM25 query provided without mode=budget — query may be silently ignored
• Payment not completed (HTTP 402) — response blocked until x402 payment fulfilled
• Malformed URL string that fails minLength or format validation