What is Aeca Search Demo with Wikipedia?
It is an interactive AI search example that utilizes Aeca's various functionalities and allows questions and answers with its own data rather than general information over the Internet. Currently, the AI search is working with Wikipedia data. You can store your own data, including merchandise, movies, books, corporate internal contracts, nutritional supplements, and job candidate profiles, in a Aeca vector database to customize interactive AI searches optimized for corporate internal data or to create internal systems.
Aeca Search Demo Real-time Statistics
Wikipedia stores newly created information or edited content within existing documents in real time in the Aeca database and provides the most up-to-date data.
- Wikipedia Change Streams for the past 24 hours:
- Total number of documents has been indexed so far:
- Total size of data stored in the database:
- Total number of queries processed:
Details of Aeca Search Demo
Aeca Search Demo has been developed using GPT 3.5 along with hundreds of millions of vector embeddings including English and Korean Wikipedia articles and public images, and it was developed on a MacBook Air M2.
- Document data coverage: Full text of English • Korean Wikidata
- Public Image data coverage: Approximately 4.3 million
- Number of Vector embeddings:
- Data size: with zstd compression
- Search functionalities: Full-Text-Search • Vector Search • Hybrid Search
- LLM integration support: Open AI GPT-3.5 and LLAMA 2
- Server specifications: Mac mini M2 • 8-core CPU • 16GB of integrated memory