# LLMLAYER API Documentation

## Docs

- [Introduction](https://docs.llmlayer.ai/introduction.md): Search, scrape, crawl, map, extract, and answer with one API for web-aware AI applications.
- [Answer API](https://docs.llmlayer.ai/answer.md): Generate web-grounded answers with /answer and real-time SSE with /answer_stream.
- [Web Search API](https://docs.llmlayer.ai/web-search.md): Get raw web, news, image, video, shopping, and scholar search results without LLM generation.
- [Scraper API](https://docs.llmlayer.ai/scrape.md): Extract a single web page as markdown, HTML, or a screenshot.
- [Extract API](https://docs.llmlayer.ai/extract.md): Extract structured data, summaries, answers, links, and brand profiles from a single web page.
- [Crawl API](https://docs.llmlayer.ai/crawl.md): Stream markdown content from multiple pages on a website.
- [Map API](https://docs.llmlayer.ai/map.md): Discover URLs on a website without scraping page content.
- [YouTube Transcript API](https://docs.llmlayer.ai/youtube-transcript.md): Extract transcripts and available metadata from public YouTube videos.
- [PDF Content API](https://docs.llmlayer.ai/scrape-pdf.md): Extract text from a publicly accessible PDF URL.
- [Models and Pricing](https://docs.llmlayer.ai/models-pricing.md): Supported model IDs and endpoint pricing for LLMLayer.
- [Rate Limits](https://docs.llmlayer.ai/rate-limits.md): LLMLayer uses tiered rate limiting based on your total credit purchases.
- [Errors & Refunds](https://docs.llmlayer.ai/errors.md): Every error format, status code, and error type the LLMLayer API returns - plus when you are automatically refunded.
- [Changelog](https://docs.llmlayer.ai/changelog.md): New features, improvements, and fixes across the LLMLayer API, SDKs, and dashboard.
- [Extract Examples](https://docs.llmlayer.ai/examples/extract.md): Ready-to-adapt Extract API recipes for structured data, lists, Q&A, summaries, and brand enrichment.
- [Search + Scrape Pipeline](https://docs.llmlayer.ai/examples/search-scrape.md): Find relevant pages, scrape full content, and generate a grounded answer.
- [Streaming Crawl Examples](https://docs.llmlayer.ai/examples/crawl.md): Handle page, usage, done, and error events from the Crawl API.
- [API Reference](https://docs.llmlayer.ai/api-reference/introduction.md): HTTP endpoints, authentication, request bodies, and generated OpenAPI schemas.
- [Answer API - Web-enhanced AI responses](https://docs.llmlayer.ai/api-reference/endpoint/answer.md): Search the web and generate AI-powered answers using your chosen LLM model. Supports multiple search queries, custom prompts, and structured JSON output.
- [Answer API - Stream responses in real-time](https://docs.llmlayer.ai/api-reference/endpoint/stream-answer.md): Stream search results and AI responses in real-time using Server-Sent Events (SSE). Does not support JSON structured output.
- [Web Search API - Raw search results](https://docs.llmlayer.ai/api-reference/endpoint/web-search.md): Search the web across multiple content types (general, news, shopping, videos, images, scholar) without AI processing. Returns raw search results.
- [Scraper API - Multi-format content extraction](https://docs.llmlayer.ai/api-reference/endpoint/scrape.md): Extract content from a non-PDF URL as markdown, HTML, or a screenshot (base64). Direct PDF URLs are rejected; use /api/v2/get_pdf_content.
- [Extract API - Multi-mode page extraction](https://docs.llmlayer.ai/api-reference/endpoint/extract.md): Extract from a single web page in one call. Combine any modes: structured JSON ('json', $0.005), a markdown summary ('summary', $0.005), an answer to a question ('qa', $0.005), all page links ('links', $0.001), and a brand profile with logos/colors/socials ('brand', $0.002). Pricing is summed per se…
- [Map API - Discover URLs on a website](https://docs.llmlayer.ai/api-reference/endpoint/map.md): Firecrawl-style URL discovery. Maps all URLs on a website with lightweight titles (no content extraction). Useful for site exploration before crawling.
- [Crawl API - Stream markdown pages from a website](https://docs.llmlayer.ai/api-reference/endpoint/crawl.md): Crawl multiple pages from a website with depth control. Streams Server-Sent Events as pages are scraped. The public endpoint currently returns markdown content per page. Billed per successfully scraped page ($0.001/page).
- [YouTube Transcript API - Extract video transcripts](https://docs.llmlayer.ai/api-reference/endpoint/youtube-transcript.md): Extract transcripts from YouTube videos with optional language selection. Returns transcript plus video metadata (title, author, views, likes, date).
- [PDF Content API - Extract text from PDFs](https://docs.llmlayer.ai/api-reference/endpoint/scrape-pdf.md): Extract text content from PDF documents. 15-second timeout with proper 504 handling.

## OpenAPI Specs

- [openapi](https://docs.llmlayer.ai/api-reference/openapi.json)

## Optional

- [Community](https://discord.gg/EqQF4cjTq5)