Vector databases are extremely useful for RAG, RecSys, computer vision, and a whole host of other ML/AI applications. Because of the rise of LLMs, there has been a lot of focus on vector indices, the…
Disclaimer: We will go into some technical and architectural details of how we do this at Neum AI A data platform for embeddings management, optimization, and synchronization at large scale…
When I first read Google’s RETRO paper, I was skeptical. Sure, RETRO models are 25x smaller than the competition, supposedly leading to HUGE savings in training and inference costs. But what about the new trillion token “retrieval database” they added to the architcture? Surely that must add back some computational costs, balancing the cosmic seesaw?