v1.9.0

Firecrawl v1.9.0 Release

What's New:

Self-Host Improvements

• Supabase client fixes
• Fixed support for LLM Providers
• Crawl is much faster now
• Global adoption of cacheable lookup system
• Easier setup

MCP Improvements (v1.11.0)

• Tons of improvements to it (prompts, examples, and how to use params properly)

SDK & API Enhancements

• Added change tracking to SDK 2.0
• Crawl delay support with per-crawl concurrency limiting
• New Qwen3 crawler example via OpenRouter
• Cancel batch scrape endpoint

Performance & Limits

• Global adoption of cacheable lookup system
• Increased map endpoint limit from 5,000 to 30,000 links
• Search schema limit increased from 50 to 100

Fixes & Stability

• Better error handling for SSL failures
• Optional chaining bug fixes
• WaitAction field validation in firecrawl-py
• Concurrency queue reworked to prioritize by time, not priority

Dashboard (Cloud version)

• New activity logs

What's Changed

• (sdk/py) Change Tracking Added to 2.0 by @nickscamara in https://github.com/mendableai/firecrawl/pull/1506
• Fix sdk/schemas by @rafaelsideguide in https://github.com/mendableai/firecrawl/pull/1507
• Fix: Handle both dict and model instances in actions parameter by @devin-ai-integration in https://github.com/mendableai/firecrawl/pull/1508
• [feat] Implement GCS storage option for scrape results across controllers an… by @rafaelsideguide in https://github.com/mendableai/firecrawl/pull/1500
• (batch/scrape) Cancel batch scrape endpoint by @nickscamara in https://github.com/mendableai/firecrawl/pull/1509
• Add Qwen3 web crawler example using OpenRouter by @devin-ai-integration in https://github.com/mendableai/firecrawl/pull/1510
• Add crawl delay functionality with per-crawl concurrency limiting (FIR-249) by @devin-ai-integration in https://github.com/mendableai/firecrawl/pull/1413
• Add delay parameter to crawl options in all SDKs by @devin-ai-integration in https://github.com/mendableai/firecrawl/pull/1514
• refactor: maximum links limit for map endpoint from 5000 to 30000 by @ftonato in https://github.com/mendableai/firecrawl/pull/1523
• fix(concurrency-limit): rework cc queue to work by time not priority (FIR-1890) by @mogery in https://github.com/mendableai/firecrawl/pull/1526
• fix(queue-worker, scrape): stealth proxy pricing change (FIR-1847) by @mogery in https://github.com/mendableai/firecrawl/pull/1521
• feat(acuc): propagate team flags (FIR-1879) by @mogery in https://github.com/mendableai/firecrawl/pull/1522
• [Bug Fix] Make WaitAction milliseconds field optional in firecrawl-py by @devin-ai-integration in https://github.com/mendableai/firecrawl/pull/1533
• create openAI provider using base url parameter by @y0hnn in https://github.com/mendableai/firecrawl/pull/1480
• refactor: increase max limit for search request schema from 50 to 100 by @ftonato in https://github.com/mendableai/firecrawl/pull/1545
• feat(scrapeURL): better error for SSL failures by @mogery in https://github.com/mendableai/firecrawl/pull/1552
• Fix/optional chaining operators missing by @tribixbite in https://github.com/mendableai/firecrawl/pull/1549
• feat(api/extract/fire-0): error logging (FIR-1971) by @mogery in https://github.com/mendableai/firecrawl/pull/1556
• feat: use cacheable lookup everywhere by @mogery in https://github.com/mendableai/firecrawl/pull/1559
• Mog/cachable lookup by @mogery in https://github.com/mendableai/firecrawl/pull/1560
• Add caching for RunPod PDF markdown results in GCS by @devin-ai-integration in https://github.com/mendableai/firecrawl/pull/1561
• Fix Supabase client configuration errors when USE_DB_AUTHENTICATION is false by @devin-ai-integration in https://github.com/mendableai/firecrawl/pull/1534

New Contributors

• @y0hnn made their first contribution in https://github.com/mendableai/firecrawl/pull/1480
• @tribixbite made their first contribution in https://github.com/mendableai/firecrawl/pull/1549

Full Changelog: https://github.com/mendableai/firecrawl/compare/v1.8.0...v1.9.0