Crawl4AI changelog: what's new each month
Every stable Crawl4AI release summarised by month: the highlights, new features, improvements, fixes and anything you need to act on. 12 months covered; the current month updates daily.
November 2026 · so far
1 release: 0.9.6
November brought a maintenance release focused on distribution and availability across package managers and container platforms.
Highlights
- Version 0.9.6 released on PyPI for direct installation via pip.
- Docker images published for containerized deployment.
October 2026
1 release: 0.9.5
October brought version 0.9.5 with updated distribution channels. The release is available through both PyPI and Docker.
September 2026
1 release: 0.9.4
September brought version 0.9.4 with updated installation options across both PyPI and Docker distributions.
Highlights
- Version 0.9.4 released for PyPI and Docker
August 2026
1 release: 0.9.3
August brought a minor release focused on distribution and availability updates. Docker images and PyPI packages were refreshed for version 0.9.3.
July 2026
2 releases: 0.9.1 → 0.9.2
July brought two minor releases focused on distribution availability. Both versions are available through PyPI and Docker.
June 2026
4 releases: 0.8.7 → 0.9.0
June saw four releases of Crawl4AI with regular updates across the month, culminating in version 0.9.0. The releases maintained consistent availability through PyPI and Docker distribution channels.
March 2026
1 release: 0.8.5
March saw the release of version 0.8.5, with Docker image availability and installation options clarified for users.
January 2026
1 release: 0.8.0
January brought version 0.8.0 of Crawl4AI with updated installation packages across PyPI and Docker platforms.
Highlights
- Version 0.8.0 released and available on PyPI
- Docker images updated and published to the registry
December 2025
1 release: 0.7.8
December brought a stable release with version 0.7.8, ensuring users can access the latest build through both PyPI and Docker.
Highlights
- Version 0.7.8 released and available on PyPI for standard pip installation
November 2025
1 release: 0.7.7
November brought production-ready monitoring and performance improvements to Crawl4AI. The release adds comprehensive observability tools and a new smart browser pool architecture designed for handling larger workloads.
Highlights
- Interactive monitoring dashboard provides real-time visibility into crawling operations.
- WebSocket streaming enables live updates during long-running crawls.
- Smart browser pool uses a 3-tier architecture for more efficient resource management.
- Async LLM extraction now supports true parallel processing without blocking.
- Janitor system automates cleanup of crawling artifacts and temporary data.
New
- Interactive monitoring dashboard accessible at /dashboard
- Comprehensive Monitor API for programmatic access to crawling metrics
- WebSocket streaming for real-time crawl updates
- Smart browser pool with 3-tier architecture
Improved
- Async LLM extraction now processes multiple requests in parallel without blocking
- CDP endpoint verification uses exponential backoff for more reliable connections
Fixed
- Async LLM extraction blocking issue that prevented true parallel processing
- CDP endpoint verification reliability with exponential backoff retry logic
October 2025
2 releases: 0.7.5 → 0.7.6
October brought expanded AI model support and webhook capabilities to Crawl4AI. The month focused on flexibility for different deployment scenarios and real-time notification workflows.
Highlights
- Multi-provider LLM support lets you use OpenAI, Gemini, or custom endpoints with configurable temperature.
- Webhooks now work with both crawl and LLM jobs, enabling real-time notifications instead of polling.
- Self-hosted model support through baseurl configuration for on-premise deployments.
- Improved Docker integration with JWT authentication fixes and stealth mode support.
- Enhanced proxy configuration with new proxyconfig structure for better parsing.
New
- Custom temperature parameter for LLM creativity control
- Multi-provider LLM support including OpenAI, Gemini, and custom endpoints
- Baseurl configuration option for self-hosted models
- Webhook support for /llm/job endpoint in addition to /crawl/job
- Custom header configuration for webhook delivery
- Global webhook URL configuration in config.yml
- Automatic retry with exponential backoff for webhook delivery
Improved
- Docker API integration enhancements
- Proxy configuration parsing with new proxyconfig structure
- Webhook payloads now include full data
- JWT authentication validation in Docker environments
Fixed
- Plus signs in URL query parameters are now preserved correctly
- Docker JWT validation issues resolved
- Playwright stealth features integration fixed
August 2025
2 releases: 0.7.3 → 0.7.4
August brought new subscription tiers for different user needs and continued refinement of the core platform.
Highlights
- New subscription plans launched: Believer, Builder, Growing Team, and Data Infrastructure Partner tiers with varying levels of support and access.
- Enhanced capability to access previously blocked corporate sites and databases.
- Improved ability to gather data from protected e-commerce and social media sources despite anti-bot measures.
New
- Believer subscription tier ($5/mo) with sponsors-only Discord access
- Builder subscription tier ($50/mo) with priority support and early feature access
- Growing Team subscription tier ($500/mo) with bi-weekly syncs and optimization assistance
- Data Infrastructure Partner tier ($2000/mo) with full partnership and dedicated support
- Access to previously blocked corporate sites and databases
- Competitor data collection from protected sources
- E-commerce pricing monitoring across anti-bot protected sites
- News and social media content collection despite protection systems
Summaries are written automatically from the official release notes (full changelog ↗); check the original notes before relying on a detail. Crawl4AI: pricing, features and alternatives · All changelogs