The community member wants to set up session management with Qdrant, LlamaIndex, and LlamaParse to avoid having to redo parsing and indexing every time. The comments suggest that they can check if the collection exists in Qdrant, and if so, they don't need to parse again. However, when loading the vector index, they need to provide nodes. The community members discuss ways to skip parsing and splitting nodes, such as using VectorStoreIndex.from_vector_store() to connect to the vector store without creating nodes. The community members provide suggestions and the original poster says they will try it out and get back to the thread if it doesn't work.
Basically, I want session management with Qdrant, LlamaIndex, and LlamaParse, so I don't have to redo parsing and indexing every time, thank you guys in advance
storage_context in this case is , vector_store = QdrantVectorStore(client=client, collection_name="test_collection") storage_context = StorageContext.from_defaults(vector_store=vector_store)
