Release Notes [2026-03-10] llama-index-core [0.14.16] Add token-bucket rate limiter for LLM and embedding API calls (#20712) Fix/20706 chonkie init doc (#20713) fix: pass tool_choice through FunctionCallingProgram (#20740) feat: Multimodal LLMReranker (#20743) feat: add optional embed_model to SemanticDoubleMergingSplitterNodeParser (#20748) fix(core): preserve doc_id in legacy_json_to_doc (#20750) fix: async retry backoff to avoid blocking event loop (#20764) Fix additionalProperties in auto-generated KG schema models (#20768) fix: respect db_schema when custom async_engine is provided (#20779) fix(core): replace blocking `run_async_tasks` with `asyncio.gather` (#20795) feat(rate_limiter): add SlidingWindowRateLimiter for strict per-minute caps (#20799) fix(core): preserve `docstore_strategy` across pipeline runs when no vector store is attached (#20824) Fix FunctionTool not respecting pydantic Field defaults (#20839) Fix MarkdownElementNodeParser to extract code blocks (#20840) security: add RestrictedUnpickler to SimpleObjectNodeMapping (CWE-502) (#20857) feat: extend vector store metadata filters (#20861) fix(react): pass system_prompt to ReActChatFormatter template (#20873)…