Crawl4AI Docker: the self-hosted crawling API

Start the container

docker run -d \
  -p 11235:11235 \
  --name crawl4ai \
  --shm-size=1g \
  unclecode/crawl4ai:latest

• API: http://localhost:11235
• Playground (test requests in the browser): http://localhost:11235/playground
• Health check: GET /health

--shm-size=1g matters: Chromium needs shared memory and crashes on larger pages without it.

Get Markdown: POST /md

curl -X POST http://localhost:11235/md \
  -H "Content-Type: application/json" \
  -d '{"url": "https://example.com", "f": "fit"}'

f selects the filter: raw, fit, bm25 (with a query q) or llm.

Full crawl: POST /crawl

curl -X POST http://localhost:11235/crawl \
  -H "Content-Type: application/json" \
  -d '{
    "urls": ["https://example.com"],
    "browser_config": {"type": "BrowserConfig", "params": {"headless": true}},
    "crawler_config": {"type": "CrawlerRunConfig", "params": {"cache_mode": "bypass"}}
  }'

Other endpoints: /crawl/stream (NDJSON), /html, /screenshot, /pdf, /execute_js, plus MCP at /mcp/sse (see Crawl4AI MCP).

LLM features: .llm.env

# .llm.env  (never commit this file)
OPENAI_API_KEY=sk-...
ANTHROPIC_API_KEY=...

docker run -d -p 11235:11235 --name crawl4ai --shm-size=1g \
  --env-file .llm.env unclecode/crawl4ai:latest

docker compose

services:
  crawl4ai:
    image: unclecode/crawl4ai:latest
    ports: ["11235:11235"]
    shm_size: "1g"
    env_file: [.llm.env]
    restart: unless-stopped

Want the request for a specific page? Enter it in the online tool and open the “Docker API” tab.