●Change8

v0.14.12

📦 llamaindex
✨ 11 features🐛 9 fixes🔧 18 symbols

Summary

This release introduces async tool spec support across several integrations, adds new LLM providers like AI Badgr and Typecast, and provides significant updates to MongoDB and Vertex AI vector stores. It also includes various bug fixes for metadata filtering and Pydantic schema handling.

Migration Steps

  1. Update google-genai to v1.52.0 if using llama-index-llms-google-genai.
  2. Note that MongoDB vector store can now initialize without list permissions.

✨ New Features

  • Added async tool spec support across multiple packages (core, agentops, azurepostgresql, lancedb).
  • Introduced Element node parser in core.
  • Added AI Badgr OpenAI-compatible LLM integration.
  • Added Typecast tool integration for text-to-speech features.
  • Added support for GPT-5.2 and GPT-5.2 Pro in OpenAI LLM.
  • Added MENTIONS edge to Nebula property graph store.
  • Added keep_alive parameter to Ollama embeddings.
  • Added support for Voyage multimodal 35.
  • Added async support and index deletion for MongoDB Atlas vector store.
  • Added Google Vertex AI Vector Search v2.0 support.
  • Added support for custom httpx.AsyncClient in BasicMcpClient.

🐛 Bug Fixes

  • Sanitized generic Pydantic model schema names for OpenAI.
  • Fixed crash in log_vector_store_query_result when result.ids is None.
  • Fixed Bedrock Converse thinking block issue.
  • Fixed missing role and improved file upload handling in Google-GenAI.
  • Fixed metadata filtering logic and SQL generation in LanceDB.
  • Fixed Chroma nested metadata filters and added multimodal result support.
  • Fixed Faiss add_with_ids by passing numpy arrays instead of integers.
  • Fixed Redis metadata filtering.
  • Handled tools=None in OpenAIResponses model kwargs.

🔧 Affected Symbols

MockFunctionCallingLLMElementNodeParserJavaCodeSplitterLiteLLMOllamaEmbeddingNebulaPropertyGraphStoreAIBadgrLLMAnthropicBedrockConverseGoogleGenAIOpenAIBasicMCPClientTypecastToolSpecChromaVectorStoreFaissVectorStoreLanceDBVectorStoreMongoDBAtlasVectorSearchRedisVectorStore