SPA Markdown Reader – BM25 Budget-Controlled Web Distillation

Choose this endpoint when you need to feed web page content to an LLM but want to minimize token consumption and cost. It is especially useful when dealing with large, noisy HTML pages where only a portion is relevant (use budget+BM25 mode), when you need structured tables extracted from documentation, or when you want a clean Markdown rendering of a SPA-rendered page. Prefer it over generic scrapers when token budget control and ROI measurement are important.

• URL is unreachable or returns a non-200 status — endpoint may return an error status
• Raw HTML is malformed and cannot be parsed — distillation may return empty or partial content
• max_tokens is set too low to capture meaningful content — excerpt may be truncated or empty
• BM25 query is too vague to meaningfully rank sections — budget excerpt may be low relevance
• SPA-rendered JavaScript content may not be fully captured if dynamic rendering fails
• Payment not completed via x402 — request rejected with 402 Payment Required