What is AI Browser?
AI Browser is a free and open-source that uses AI to do things for you, like search the web or manage files. You pick your AI provider in the settings, then chat with it or let it handle complex tasks automatically. It’s built for learning, so it works best on Mac and might need a few tries to build on Windows.
Features
- Dual Mode Interaction:
- Chat Mode (default): Lightweight conversation with ChatAgent, streaming text, tool calls (web search, webpage QA), thinking display
- Deep Explore Mode: Multi-step workflow generation → user confirm → multi-agent execution for complex tasks
- Web Search Integration: Built-in web search via Tavily, Serper, or SearXNG (self-hosted) with configurable search provider
- Multiple AI Providers: Support for DeepSeek, Qwen, Google Gemini, Anthropic Claude, and OpenRouter
- Complete Settings System: Configure everything through the UI, no file editing required
- Providers: API keys, models, and provider-specific settings
- General: Language, startup behavior, window preferences
- Chat: Temperature, max tokens, streaming, search provider configuration
- Agent: Browser/File/Custom agent settings, per-agent MCP services
- UI: Theme (Dark/Light/System), font size, density, editor preferences
- Network: Proxy, request timeout, stream timeout, retry attempts
- Workflow Control: Editable workflow steps with pause/resume, expert mode, and thinking display
- Scheduled Tasks: Create and manage automated recurring tasks with custom intervals
- AI-Powered Browser: Intelligent multi-tab browser with automated task execution
- Multi-Modal AI: Vision and text processing capabilities
- Speech & TTS: Voice recognition and text-to-speech integration
- File Management: Advanced file operations and management
- Internationalization: Full English and Chinese language support
- AI-powered browser with automated task execution
- Multiple AI provider support (DeepSeek, Qwen, Google Gemini, Claude, OpenRouter)
- Multi-modal AI capabilities (vision and text processing)
- Scheduled tasks system with custom intervals
- File management capabilities
- UI configuration for API keys and models
- Human interaction support (AI can ask questions during execution)
- Task continuation with file attachment management
- Atomic fragment-based history playback with typewriter effects
- Advanced playback controls (play/pause/restart/speed adjustment)
- Context restoration and session management
- Optimized auto-scroll behavior for messages
- Enhanced message display and rendering
- Chat mode with ChatAgent (streaming text + inline tool calls)
- Deep Explore mode (renamed from workflow mode) for complex multi-step tasks
- Web search integration with 3 providers (Tavily, Serper, SearXNG)
- Webpage QA — extract and analyze content from open browser tabs
- Editable workflow steps with user confirmation before execution
- Task pause/resume support
- Expert mode with thinking process display
- Per-agent MCP service configuration
- Custom agent support with user-defined prompts
- Multi-window task isolation and global service injection
