Kagi

Kagi is a subscription-based search engine and web ecosystem designed to prioritize user privacy, control, and ad-free browsing. Unlike conventional search engines that rely on advertising revenue and track user behavior to personalize results, Kagi operates on a direct payment model, meaning users pay for the service instead of being the product. The core offering, Kagi Search, provides search results devoid of advertisements, sponsored links, and promotional clutter. Users have granular customization options: they can adjust the weight of different sources in search rankings, block specific domains entirely, promote preferred websites, and apply community-created filters called Lenses (e.g., for news, academic content, or programming). These controls allow individuals to shape their search experience around their own preferences rather than algorithms optimized for ad engagement. Privacy is a fundamental feature: Kagi does not track clicks, store search history tied to user identities, or share data with third parties. All queries are anonymized, and the platform uses a secure, encrypted connection by default. Beyond search, Kagi offers the Kagi Assistant, a privacy-respecting interface to multiple large language models (LLMs) such as OpenAI’s GPT-4, Anthropic’s Claude, and Meta’s LLaMA. The Assistant does not log queries or use them to train the underlying models; each response is grounded in Kagi’s search index, ensuring factual accuracy and reducing hallucinations. Users can compare outputs from different models side-by-side within the same session, choose preferred models for different tasks, and configure system prompts without fear of data leakage. The Assistant also supports file uploads (images, PDFs, text) for analysis, making it useful for research, summariz