IntelliSheet AI
Turn spreadsheets into intelligent data engines with native LLM orchestration and automated formula generation.
The privacy-first browser AI assistant for seamless content synthesis and context-aware automation.
MeetLeo (Leo) is the native AI assistant developed by Brave Software, integrated directly into the Brave browser architecture. Unlike standard extensions, Leo operates with a privacy-first engineering focus, utilizing an anonymized proxy server to ensure that user prompts are never linked to IP addresses or stored for model training. By 2026, Leo has evolved into a multi-model orchestrator, allowing users to toggle between Llama 3, Mixtral 8x7B, and Claude 3.5 Sonnet depending on the task complexity. Technically, it leverages Retrieval-Augmented Generation (RAG) to process the active Document Object Model (DOM) of the current tab, enabling high-fidelity summarization, real-time translation, and code analysis without data leaving the local context in an identifiable way. Its market position is unique as it bypasses the friction of third-party logins, providing immediate utility to over 60 million Brave users. For developers and researchers, Leo offers a sandbox-like environment to interact with web data through natural language, effectively turning every URL into a queryable database while maintaining a zero-knowledge security posture.
Directly reads and indexes the active browser tab's DOM to provide answers grounded in current content.
Turn spreadsheets into intelligent data engines with native LLM orchestration and automated formula generation.
The permission-aware generative AI and search platform for enterprise knowledge.
A personal AI designed to be supportive, smart, and available anytime.
AI-powered PDF management for smarter document analysis and seamless workflows.
Verified feedback from the global deployment network.
Post queries, share implementation strategies, and help other users.
Routes all LLM requests through a proxy server that strips identifiable metadata and IP headers.
Allows hot-swapping between Anthropic and Meta models within the same chat session.
Automatically parses local PDF files opened in-browser as context for the assistant.
Generates citations with clickable timestamps/links for the information extracted from the page.
Ensures that the server-side LLM provider does not retain logs for training purposes.
Uses Brave Search API to fetch live data beyond the existing page context.
Developers spending hours reading long READMEs to find specific implementation details.
Registry Updated:2/7/2026
Information overload when reviewing news or lengthy articles.
Analyzing competitor sites written in foreign languages.