v0.13.4
📦 llamaindexView on GitHub →
✨ 10 features🐛 3 fixes⚡ 1 deprecations🔧 12 symbols
Summary
This release introduces Baseten and ZenRows integrations, adds MMR search to Chroma, and updates IBM integrations with external URL support while deprecating instance_id. It also includes critical fixes for ContextChatEngine and Postgres vector store deletion logic.
Migration Steps
- Update IBM integration configurations to use external URLs instead of 'instance_id'.
- If using Postgres vector store deletion, ensure 'ref_doc_id' is correctly indexed as the logic has shifted from 'doc_id'.
✨ New Features
- Add PostgreSQL schema support to Memory and SQLAlchemyChatStore.
- Add sync wrapper for put_messages in memory modules.
- Add initial tool choice option in FunctionAgent.
- New integration: Baseten (Embeddings and LLMs).
- Support for additional/external URLs in IBM integrations.
- Add amazon.nova-premier-v1:0 support to Bedrock Converse.
- Raise Google Drive errors in Google readers.
- New ZenRows web reader integration.
- Add MMR (Maximal Marginal Relevance) search to Chroma vector store.
- Add Payload indexes support to QdrantVectorStore.
🐛 Bug Fixes
- Fix ContextChatEngine to correctly handle QueryBundle objects instead of just strings.
- Fix SentenceTransformerRerank initialization device handling.
- Fix Postgres vector store deletion to use indexed 'ref_doc_id' instead of 'doc_id'.
🔧 Affected Symbols
MemorySQLAlchemyChatStoreFunctionAgentContextChatEngineSentenceTransformerRerankChromaVectorStoreQdrantVectorStoreZenRowsReaderIBMEmbeddingIBMLLMIBMRerankPGVectorStore⚡ Deprecations
- The 'instance_id' parameter is deprecated in IBM integrations (LLMs, Embeddings, Postprocessor) in favor of external URL support.