Perplexica

Perplexica is a privacy-conscious AI search and answer engine that runs natively on personal hardware. It integrates knowledge from the Internet, supports local LLMs (such as Olama) and cloud AI service providers (such as OpenAI, Claude, Groq), Be able to provide accurate, cited answers while ensuring that the search process is completely private. [Core Features] 1. Support major AI providers * Support local LLM and multiple cloud AI models * Different models can be mixed and called according to needs, making it flexible and efficient 2. Intelligent search model * Balancing mode: Balancing speed and quality * Fast Mode: Priority response speed * Quality Model (coming soon): Pursuing the highest quality of answers 3. Focus Modes * Academic paper model * YouTube Video Mode * Reddit discussion model * Wolfram Alpha Computing Model * Writing assistant model * Universal web search model 4. SearxNG-based privacy search * Aggregate multiple search engine results * Protect user identity and privacy * Advanced search engines such as Tavily and Exa will be supported in the future 5. Image and video search * Provide visual search results (images, videos) * No longer limited to plain text answers 6. File upload and Q & A * Support uploading documents and conducting content Q & A * Compatible formats include PDF, text files, etc. 7. Restricted domain search * You can limit your search to specific websites * Suitable for technical documents or academic research scenarios 8. smart recommendations * Provide real-time smart prompts when entering search content * Help users build more accurate queries 10. search history * All search records are stored locally * Users can review and manage their exploration history at any time