1 Release published by 1 person

• v0.12.19

  published Feb 18, 2025

9 Pull requests merged by 7 people

• fix: Chat messages with tool calls incorrectly mapping to Vertex message

  #17893 merged Feb 24, 2025

• fix: FunctionAgent and ReActAgent error out when LLM response contains no chat messages

  #17884 merged Feb 22, 2025

• Path was incorrect for microsoft_outlook_emails integration

  #17867 merged Feb 20, 2025

• attach more metadata to retrieved image node

  #17868 merged Feb 20, 2025

• fix: bump python-jose

  #17866 merged Feb 19, 2025

• v0.12.19

  #17848 merged Feb 18, 2025

• Experimental natural language retrievers using duck db

  #15642 merged Feb 17, 2025

• Function tool callback

  #16637 merged Feb 17, 2025

• mistral saba support

  #17847 merged Feb 17, 2025

15 Pull requests opened by 15 people

• fix ChatSummaryMemoryBuffer._summarize_oldest_chat_history

  #17845 opened Feb 17, 2025

• Refactor from_defaults to Use tool_metadata_from_defaults for Cleaner and More Efficient Metadata Handling

  #17852 opened Feb 18, 2025

• Support text-only prompts for LlamaMultiModal class

  #17855 opened Feb 18, 2025

• Add license files to each of the subpackages

  #17856 opened Feb 19, 2025

• feat: auto-detect custom start and stop events in workflow classes

  #17865 opened Feb 19, 2025

• tl;dw AI Integration for Retrievers

  #17872 opened Feb 20, 2025

• feat: change maximum allowed model temperature to 2.0 for Gemini

  #17886 opened Feb 21, 2025

• fix: Convert Embedding Functions to Async

  #17888 opened Feb 22, 2025

• feat: add tests for gemini, fix function calling parameters

  #17889 opened Feb 22, 2025

• fix: fix adelete method and add delete_nodes to elasticsearch vector store

  #17890 opened Feb 22, 2025

• feat: Add Valyu Integration

  #17892 opened Feb 22, 2025

• Fix CitationQueryEngine init function for response_synthesizer

  #17897 opened Feb 24, 2025

• feat: Add `ibm-watsonx-ai` rerank integration

  #17900 opened Feb 24, 2025

• only modify delta if 'Answer:' was actually detected

  #17901 opened Feb 24, 2025

• update for redisvl 0.4.0

  #17902 opened Feb 24, 2025

9 Issues closed by 10 people

• [Question]: Request ID tracing in the new instrumentation module

  #17894 closed Feb 24, 2025

• WorkflowRuntimeError: Error in step 'run_agent_step'

  #17891 closed Feb 22, 2025

• [Documentation]: update nvidia multi modal notebook to mention NVIDIAMultiModal

  #17478 closed Feb 22, 2025

• [Bug]: ModuleNotFoundError: No module named 'llama_index'

  #15749 closed Feb 20, 2025

• StepDecomposeQueryTransform "NOTE: doesn't work yet."

  #17876 closed Feb 20, 2025

• [Question]: Is it possible to implement cyclic graph with llamaindex workflow?

  #17874 closed Feb 20, 2025

• [Bug]: Function name misplace with fn name

  #17860 closed Feb 20, 2025

• [Bug]: Metadata filtering does not filter at all

  #17857 closed Feb 20, 2025

• [Question]: Workflow Context, Global or Local?

  #17854 closed Feb 18, 2025

28 Issues opened by 26 people

• [Bug]: KeyError: 'commit' in GithubRepositoryReader

  #17903 opened Feb 24, 2025

• [Bug]: Llama Index Workflow: step-wise does not work

  #17898 opened Feb 24, 2025

• [Bug]: CitationQueryEngine init function

  #17896 opened Feb 24, 2025

• [Feature Request]: Bedrock prompt caching

  #17895 opened Feb 24, 2025

• [Bug]: Unable to load FaissIndex Vector Store

  #17887 opened Feb 22, 2025

• [Bug]: Not able to use chroma db vector store integration to get nodes using get_nodes

  #17885 opened Feb 21, 2025

• [Question]: Vector PG Retriever returns triplets that do not appear in the actual graph

  #17883 opened Feb 21, 2025

• [Bug]: Can't provide images to Gemini models

  #17882 opened Feb 21, 2025

• [Question]: Complex state update

  #17881 opened Feb 21, 2025

• [Bug]: is_done for workflows will always return True

  #17880 opened Feb 21, 2025

• [Feature Request]: Make use of PyPI's trusted publishing

  #17879 opened Feb 20, 2025

• [Feature Request]: Make use of GitHub dependency-review

  #17878 opened Feb 20, 2025

• [Bug]: async query of Postgres-based RetrieverQueryEngine produces syntax error

  #17877 opened Feb 20, 2025

• [Bug]: siliconflow can not stream response

  #17873 opened Feb 20, 2025

• [Bug]: Hash of Node does not consider metadata

  #17871 opened Feb 20, 2025

• [Question]: 【llamaindex multimodal propertygraphindex neo4j 】

  #17870 opened Feb 20, 2025

• [Bug]: Function name misplace with fn name

  #17869 opened Feb 20, 2025

• [Bug]：模块导入

  #17864 opened Feb 19, 2025

• [Question]: llama-index-core dependencies

  #17863 opened Feb 19, 2025

• [Bug]: temperature and other llm argument used in tree_summarize

  #17862 opened Feb 19, 2025

• [Question]: How can I run a correctness evaluator using Azure OpenAI?

  #17861 opened Feb 19, 2025

• [Bug]: Xorbits Inference RESTfulChatModelHandle.chat() got an unexpected keyword argument 'prompt'

  #17859 opened Feb 19, 2025

• [Bug]: Tool description exceeds maximum length of 1024 characters. Please shorten your description or move it to the prompt.

  #17858 opened Feb 19, 2025

• [Bug]: chromadb.AsyncHttpClient &&& ChromaVectorStore : 'coroutine' object is not subscriptable

  #17853 opened Feb 18, 2025

• [Bug]: using fastapi and ElasticsearchStore will result in an error: RuntimeError: There is no current event loop in thread 'AnyIO worker thread'.

  #17851 opened Feb 18, 2025

• [Bug]: llama_index.readers.discord

  #17850 opened Feb 18, 2025

• [Question]: about metadatafilter function: to filter empty array

  #17849 opened Feb 18, 2025

• [Bug]: Encoding error in the case which use Japanese texts in simple-labelled.py

  #17846 opened Feb 17, 2025

10 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• added JSONReader to file mappings

  #14419 commented on Feb 17, 2025 • 1 new comment

• [Question]: RAG implementation from existing knowledge graph

  #17810 commented on Feb 17, 2025 • 0 new comments

• [Question]: Is there any way to support the deployment of a local embedding API and a reranker API?

  #15303 commented on Feb 18, 2025 • 0 new comments

• [Feature Request]: Openai_like embedding model integration

  #14340 commented on Feb 18, 2025 • 0 new comments

• [Question]: All data from all user sessions going into Chatengine except for specific session data

  #16059 commented on Feb 18, 2025 • 0 new comments

• [Bug]: Incorrect Output in ReAct Mode of LlamaIndex Chat Engine

  #17322 commented on Feb 20, 2025 • 0 new comments

• [ReAct][Robustness++] Avoid "Reached max iterations" errors by telling Agent how many steps are allowed

  #12426 commented on Feb 17, 2025 • 0 new comments

• Add missing fields to the returned nodes from DatabricksVectorSearch

  #15947 commented on Feb 17, 2025 • 0 new comments

• Fix: Proper WMF/EMF image handling in PPTXReader

  #17819 commented on Feb 17, 2025 • 0 new comments

• Update `WatsonxLLM.metadata` property to avoid validation error when `model_limits` field isn't present

  #17839 commented on Feb 21, 2025 • 0 new comments