List view
- No due date•0/2 issues closed
- No due date•2/7 issues closed
- No due date•10/10 issues closed
initial release will deliver the core functionality essential to a robust Retrieval-Augmented Generation (RAG) pipeline. Here’s what’s included: ## 1. Ingest ### Data Connectivity: - Pre-built integrations for popular data sources such as Google Drive. Automatic syncing and file upload via simple API calls. #### Objective: Seamlessly connect your application to user data, making ingestion straightforward and reliable. ## 2. Chunk and Index ### Data Preparation: - Automatic chunking of large documents to optimize processing for language models. - Generation of embeddings using the latest multi-lingual LLMs. ### Indexing: - Build and store vector, summary, and keyword indexes in a scalable vector database. #### Objective: Transform raw data into a structured, searchable format, ensuring efficient retrieval. ## 3. Retrieve ### Advanced Retrieval API: - Semantic search capabilities that deliver the most relevant document chunks. - Built-in advanced features such as LLM re-ranking, summary indexing, entity extraction, and flexible filtering. - Hybrid search support combining semantic and keyword-based approaches. #### Objective: Provide accurate, context-aware results for any query, powering your AI applications with high-quality data retrieval.
No due date•6/6 issues closed