Paywall-Clean URL & HTML Distiller with BM25 Budget Control

Choose this endpoint when you need to feed webpage content to an LLM but want to minimize token consumption and cost — especially useful for long documentation pages, paywalled articles, or any HTML where only the topically relevant sections matter. The BM25 budget mode is uniquely suited for RAG pipelines where you need only the highest-relevance passages within a strict token ceiling. Prefer this over generic scraping endpoints when ROI proof (tokens saved, cost saved) is important for auditing or optimization workflows.

• URL is unreachable or returns non-200 → fetch error with status detail
• Paywall blocks content fetch → partial or empty distillation
• max_tokens set too low for meaningful content → truncated output or error
• Invalid mode enum value → 400 validation error
• HTML input is malformed or empty → empty or degraded distillation
• BM25 query not provided when mode=budget → fallback or error
• Rate limit or payment failure via x402 → 402 Payment Required