search

deepset-ai / haystack

AI orchestration framework to build customizable, production-ready LLM applications. Connect components (models, vector DBs, file converters) to pipelines or agents that can interact with your data. With advanced retrieval methods, it's best suited for building RAG, question answering, semantic search or conversational agent chatbots.
https://haystack.deepset.ai
Apache License 2.0
17.72k stars 1.92k forks source link

Hosted vector stores: Vertex Search AI, AWS Knowledge Bases, Azure AI Search #8290

Closed scosman closed 2 months ago

scosman commented 2 months ago

I'm curious if there's a reason you've stayed away from the big-tech vector/doc search tools like Google Vertex Search AI, AWS Knowledge Bases, Azure AI Search.

Don't get me wrong: I love pgvector, etc. But the ease of use of 100% hosted services is sometimes helpful.

Describe the solution you'd like Any chance you're adding these, or is langchain/llamaindex more appropriate for these abstractions.

silvanocerza commented 2 months ago

We have limited resources, both in terms of people and time, so we tend to focus first on integrations that people ask. We risk spreading ourselves too thin if we have too many integrations to maintain too.

As an example we have a tracking issue for Azure AI Search and are actively working on it cause the community asked for that.

If anyone feels the need to ask for a new integration they can easily ask by creating a new issue in the core integrations repo. :)

Or if they want they can also create and maintain their own integration and we'll feature it in our integration page.

scosman commented 2 months ago

Got it. Wasn't sure if it was philosophical or just priority. Thanks and makes sense.

Feel free to close if you don't find this issue helpful for tracking.