bobvanluijt

Super interesting - might be an opportunity for a Weaviate module as well (Weaviate modules take care of vectorization but are model agnostic)

Correct - this is actually a core focus of Weaviate. That's why it's not limited to only ANN but also has an inverted index, optional modules, etc.

Somebody already mentioned it on this thread, but with Weaviate, we aim to create more end-to-end experience that can be used open source or as (hybrid-)SaaS. For example, you can store the complete data object, ANN and inverted filters (hybrid search), and optional modules for vectorization. More info about the concept behind Weaviate here: https://weaviate.io/developers/weaviate/concepts

Previous post: https://weaviate.io/blog/ann-algorithms-hnsw-pq

Thanks, Jason! Very soon, the public beta of the cloud version will be live with a brand new UI, and 100% self-service hosted Weaviates

Previous article: https://weaviate.io/blog/ann-algorithms-vamana-vs-hnsw

Yeah - it will become available as a multi2vec Weaviate module as well in due time.

I'm affiliated with Weaviate and am super happy to see the role of vector databases mentioned in the report (p.15 & p.22)

Lots of things are happening in the world of vector DBs. For those interested in what Weaviate will do this year. Tl;Dr: ingestion and search pipelines, native roaring bitmap index, auto-scaling, product quantization, and disk-based ANN.

Thanks! Def looking at OOTB Lang-Chain & Weaviate possibilities.

We've just released (open source) a Weaviate module that allows you to pipe search results directly through an LLM (in this case, OpenAI's GPT-3, soon more to come). Would love to hear what you all think about this approach. More context and examples here: https://twitter.com/bobvanluijt/status/1623041783402795044

Tomorrow https://weaviate.io will also release a generative search module that accepts LaMDA when it’s in GA besides GPT-based models. (I'm affiliated with Weaviate)