Mayank

The collection is not recreated but, every time you run, `Qdrant.from_documents(...)` it will re-create embeddings. If you want to reuse your collection, create a Qdrant client like this, in a...

@a172166 more example ```from qdrant_client import QdrantClient, models from langchain_community.vectorstores import Qdrant from langchain_cohere import CohereEmbeddings client = QdrantClient(path="local_qdrant") embeddings = CohereEmbeddings(model="embed-english-light-v3.0", cohere_api_key = "") collection_name = "my_documents" qdrant =...

Please check if it is relevant to add this feature @hwchase17 @baskaryan

Hey @Bennoo since you are using AzureOpenAI's GPT-4o and the error itself looks like JSON confinement. It is recommended to add an additional argument of `response_format = {"type": "json_object"}`. ```python...

Hi @koi646 Thank you for pointing this out. The latest version of mem0ai fixes the above issue.

Hi @tslmy Currently we don't support `OpenAIAgent` on llama-index. We will increase support to more agents. Please refer this documentation: https://docs.mem0.ai/integrations/llama-index

Hi @Dev-Khant Let me know if I want to change prompt or add something.

Hi @deshraj @prateekchhikara Let me know if we need any improvements. Will work on TCs post finalisations of prompts and flow.

Hi @deshraj @prateekchhikara, I have done the validation using pydantic for `CustomCategories`. The changes seems better. Please review. Thanks

@kritinv As a user if I want to use custom embedding model, I am prompted to re-initialize the LLM in the `critic_model` argument. The expectation is that it should take...