Semantic HTML Pruner with BM25 Budget Control

Choose this endpoint when you need to reduce the token footprint of web content before passing it to an LLM, especially when working with large documentation pages, e-commerce sites, or any URL where raw HTML would exceed context limits. It is particularly valuable when you have a specific topic focus (use budget+BM25 mode) or need structured tables and metadata extracted (use structured mode). Prefer it over generic scrapers when token cost savings and LLM-ready output format are priorities.

• URL is unreachable or returns non-200 status — endpoint may return an error status
• HTML payload is empty or malformed — distillation may produce empty or low-quality output
• BM25 query not provided when mode=budget — may default to unfiltered content
• max_tokens ceiling too low (below 20) — rejected by schema validation
• Rate limiting or payment failure via x402 — billing object absent, request rejected
• Timeout fetching a slow or large URL — partial content or timeout error returned