Perplexica's architecture consists of the following key components:
- User Interface: A web-based interface that allows users to interact with Perplexica for searching images, videos, and much more.
- Agent/Chains: These components predict Perplexica's next actions, understand user queries, and decide whether a web search is necessary.
- SearXNG: A metadata search engine used by Perplexica to search the web for sources.
- LLMs (Large Language Models): Utilized by agents and chains for tasks like understanding content, writing responses, and citing sources. Examples include Claude, GPTs, etc.
- Embedding Models: To improve the accuracy of search results, embedding models re-rank the results using similarity search algorithms such as cosine similarity and dot product distance.
- Web Content
- In Agent mode, the application uses an agentic workflow to answer complex multi-part questions
- The agent can use reasoning steps to provide comprehensive answers to complex questions
- Agent mode is experimental and may consume lots of tokens and take a long time to produce responses
- In Speed mode, the application only uses the preview content returned by SearXNG
- This content is provided by the search engines and contains minimal context from the actual web page
- This mode prioritizes quick responses over accuracy
- In Agent mode, the application uses an agentic workflow to answer complex multi-part questions
For a more detailed explanation of how these components work together, see WORKING.md.