SEARCHSearch & retrieval
Jev Reverse XKCD Lookup
Hemanth.HM@GNUmanth
0:41This project replaces a Gemini Multimodal Embeddings, ChromaDB, and Cosine similarity-based reverse XKCD lookup app with a Jev-based system that eliminates embeddings, vector databases, and pre-indexing, achieving approximately 120ms latency and calibrated probabilities.
So, I had this relatively old app that did reverse @xkcd lookup... that used Gemini Multimodal Embeddings + ChromaDB + Cosine similarity.
Swapped @typesafeai's jev:
- 0 embeddings
- 0 vector database
- 0 pre-indexing
- ~120ms latency
- Calibrated probabilities
🤯 x.com/GNUmanth/status/2101808956180217886/
Related builds
Vectorless RAG with Jev
This project introduces a novel approach to Retrieval Augmented Generation (RAG) that replaces traditional embedding methods with a structured, table-of-contents-based navigation system, significantly improving speed and reducing costs by offloading non-generative tasks to Jev.
