Paywall-Clean Web Distiller with BM25 Budget Control

Choose this endpoint when you need to extract clean, token-efficient content from a URL or HTML blob for downstream LLM processing, especially when context window size or LLM cost is a concern. It is particularly valuable when you want BM25-ranked relevance filtering to surface only the most query-relevant passages within a strict token ceiling. Prefer it over generic web scrapers when you need structured JSON tables, skeleton outlines, or Markdown alongside token savings metrics. It supports both live URL fetching and offline raw HTML distillation, making it versatile for both real-time pipelines and pre-fetched content workflows.

• URL is unreachable or returns a non-200 status — endpoint may return an error status
• HTML payload is empty or malformed — distillation may fail or return minimal output
• max_tokens is set too low relative to content size — output may be aggressively truncated
• BM25 query is missing when mode=budget — may fall back to default behavior or return an error
• Payment not provided or insufficient — returns HTTP 402 requiring x402 USDC payment
• Rate limiting or upstream paywall blocks fetch — content may be unavailable