Ghita here, CEO and Co-Founder of ZeroEntropy (YC W25).
We built ZeroEntropy to help developers deploy more accurate retrieval systems, faster. Using our API, you can upload documents of any type, and retrieve accurate and relevant information from your knowledge base, in just a few lines of code.
We just released a new open-weight reranker that outperforms models like Cohere's rerank-3.5, or even Gemini Flash used as a reranker. You can check out the performance here.
Great work team! Reranker was smooth to integrate and has drastically improved our AI agent's accuracy!
this looks like something I could actually use. Does it work out of the box with existing vector DBs like Pinecone or Weaviate, or does it require its own setup?
Theoretically I can guess at the likely positive results but theres an upfront friction to witness those results and I'm not sure of the ROI. It would be good to see a side by side comparison, The POST requests in the docs are good but I imagine something even easier to show the comparative value would be very compelling.
Really impressed by how ZeroEntropy takes AI search to a whole new level—retrieval feels almost human now, kudos to the team for pulling this off so seemlesly!
