Wrap the DeepLake vector store in a retrieval system that can translate natural language queries to DeepLake-compatible structured queries, including metadata filters.
LocalAI lets your run LLMs, embedding models and others locally. And its drop-in replacement interface for OpenAI makes it super easy to get off the ground.
Use LocalAI embeddings in π¦π thanks to @mudler_it!
We've been working hard to make our docs better. This is an ongoing effort, so if you have specific feedback or requests, please let us know.
Here's what we've done in the past week:
a π§΅
1/n
Unified JS and Python structure
We've unified the structured of the JS and Python docs, so that they broadly share the same terms and organization, making it clear where there's parity.
We've moved our Use Cases and Integrations sections to the top level and flattened them, to cleanly separate conceptual and more applied docs and make it easier to get to the page you're looking for.
Predibase enables anyone to train their own models. With the new π¦π integration, it's now also easy for anyone to take their custom model and start building applications on top of it πͺ
π₯ Query @Golden knowledge base
βοΈ Search flights with @AmadeusITGroup
π Interact with geospatial data vs @geopandas
ποΈ Caching, tracing, tagging, retries, and more with @PortkeyAI
a π§΅:
π₯ Query @Golden knowledge base
Golden is aiming to build the largest open knowledge graph of entities and topics. Now your agents and chains can access that knowledge.
A unified platform to help developers debug, test, evaluate, and monitor their LLM applications.
Integrates seamlessly with LangChain, but doesn't require it.
βοΈBlog
what we built, where weβre going, and how our Alpha partners put LangSmith to use
ππ½ to companies like @klarna @SnowflakeDB, @streamlit @BCG @DeepLearningAI_ @fintual @mendableai @multion_ai & @quivr_brain for helping us shape LangSmith
4β£ GPT4All embeddings
ποΈAsync support (and more) for @qdrant_engine vecstore
π§ Tongyi Qianwen LLM
Let's take a look π§΅
ποΈAsync support (and more) for @qdrant_engine vecstore
@LukawskiKacper's latest contribution adds full async support along with MMR search and deletion capabilities to the already very capable Qdrant vector store interface.