1 Release published by 1 person

• v0.12.23

  published Mar 7, 2025

35 Pull requests merged by 21 people

• fix bugs and optimize extraction prompt and parse_fn

  #18046 merged Mar 12, 2025

• Update Vectara Index to return both document level and page level metadata

  #17976 merged Mar 12, 2025

• fix(multimodal): check image path is file before open

  #18043 merged Mar 12, 2025

• Use int type for redirect URI port

  #18097 merged Mar 12, 2025

• Fix the redundancy of ElasticsearchReader.

  #18106 merged Mar 12, 2025

• Fix typo in comment of LlamaIndex model guide (multi-modal code)

  #18103 merged Mar 12, 2025

• fix null model meta references

  #18109 merged Mar 12, 2025

• Contextual Generate model

  #17913 merged Mar 12, 2025

• Contextual reranker

  #18075 merged Mar 12, 2025

• fix gemini roles

  #18108 merged Mar 12, 2025

• - fix: rssReader support customized user-agent

  #18076 merged Mar 10, 2025

• improve MCP README

  #18080 merged Mar 10, 2025

• remove

  #18081 merged Mar 10, 2025

• Google genai embeddings

  #18079 merged Mar 10, 2025

• - Mem0Memory Integration Issue: Update Required for Mem0 API Parameter Changes

  #18066 merged Mar 10, 2025

• Fix the TypeError in the upsert_triplet method of the NeptuneBaseGrap…

  #18051 merged Mar 9, 2025

• [BUG FIX] Google genai vertexai error

  #18070 merged Mar 9, 2025

• Update workflow example to include the stepwise=True parameter

  #18017 merged Mar 9, 2025

• fix link to llama_deploy doc

  #18065 merged Mar 9, 2025

• Logan/readme update for genai

  #18059 merged Mar 8, 2025

• Use GenAI package for google

  #17939 merged Mar 8, 2025

• Update mongo as supporting hybrid search

  #18052 merged Mar 7, 2025

• v0.12.23

  #18050 merged Mar 7, 2025

• Change AgentWorkflow to FunctionAgent in documentation.

  #18042 merged Mar 7, 2025

• chore: bump jinja version in docs dependencies

  #18047 merged Mar 7, 2025

• added merging_separator argument to allow for specifying chunk merge

  #18027 merged Mar 6, 2025

• Fix the error raised when ReactAgent is created without an explicit system message

  #18041 merged Mar 6, 2025

• Add support for running single-agent workflows within the BaseWorkflowAgent class

  #18038 merged Mar 6, 2025

• chore: skip deeplake tests when running on CI

  #18015 merged Mar 6, 2025

• build(deps-dev): bump jinja2 from 3.1.5 to 3.1.6

  #18024 merged Mar 6, 2025

• build(deps-dev): bump jinja2 from 3.1.5 to 3.1.6 in /llama-index-core

  #18025 merged Mar 6, 2025

• AzureOpenAI: api_base and azure_endpoint are mutually exclusive

  #18037 merged Mar 6, 2025

• Remove base url validation

  #18031 merged Mar 6, 2025

• docs: document custom start/stop events

  #18034 merged Mar 6, 2025

• anthropic caching and thinking updates

  #18039 merged Mar 6, 2025

14 Pull requests opened by 12 people

• Add new apac bedrock llms

  #18045 opened Mar 7, 2025

• Ensuring original text is preserved in CHUNKING_REGEX

  #18054 opened Mar 8, 2025

• feat(mongo reader): field_extractors

  #18063 opened Mar 8, 2025

• Update UA of AzureCosmosDBMongoDBVectorSearch

  #18074 opened Mar 10, 2025

• add netmind integrations

  #18078 opened Mar 10, 2025

• fix: make run_step return all the events produced by a step

  #18082 opened Mar 10, 2025

• feat: Add TTL support for DynamoDB chat store

  #18084 opened Mar 10, 2025

• Add support for providing optional logits processors to HuggingFace MLLM models for output control

  #18086 opened Mar 10, 2025

• AIMon reranker integration into LlamaIndex node postprocessors

  #18087 opened Mar 10, 2025

• AIMonRerank.ipynb notebook added to LlamaIndex docs/examples

  #18088 opened Mar 11, 2025

• fix get content node metadata

  #18110 opened Mar 12, 2025

• remove assert for multimodal llms

  #18112 opened Mar 12, 2025

• bump compatible pinecone version

  #18113 opened Mar 12, 2025

• Add build_semantic_nodes_from_nodes to SemanticDoubleMergingSplitterNodeParser

  #18114 opened Mar 12, 2025

23 Issues closed by 8 people

• [Bug]: Gemini models break async chats with a ReActAgent

  #17796 closed Mar 12, 2025

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

  #17903 closed Mar 12, 2025

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

  #17885 closed Mar 12, 2025

• [Bug]: loading data using ElasticsearchReader will result in redundancy.

  #18102 closed Mar 12, 2025

• [Bug]: AzureOpenAI abstract method _prepare_chat_with_tools

  #17948 closed Mar 12, 2025

• [Bug]: QdrantVectorStore with IngestionPipeline arun 'NoneType' object has no attribute 'collection_exists'

  #18105 closed Mar 12, 2025

• [Documentation]: transformations

  #18093 closed Mar 12, 2025

• [Bug]: Similarity caculation of TextNode with or not with metadata is different.

  #18092 closed Mar 11, 2025

• [Bug]: LamaIndex.core is not found error when creating multimodal vector store

  #18091 closed Mar 11, 2025

• [Feature Request]: Integration of Atomstate (a vector store) with LlamaIndex for Enhanced Data Retrieval and Management

  #16547 closed Mar 10, 2025

• [Question]: "Score" in NodeWithScore

  #18077 closed Mar 10, 2025

• [Bug]:

  #18073 closed Mar 10, 2025

• [Bug]: TypeError: Incorrect String Formatting in `upsert_triplet` Method of NeptuneBaseGraphStore

  #18044 closed Mar 9, 2025

• [Bug]: QdrantVectorStore Fails in LlamaIndex Due to Deprecated collection_exists() Method

  #18061 closed Mar 9, 2025

• [Feature Request]: Collect events in workflow context with predicate

  #16529 closed Mar 9, 2025

• [Feature Request]: Allow Custom Suffix for Metadata Filters in Elasticsearch Index

  #16518 closed Mar 8, 2025

• [Question]: How to handle same word in query and nodes but have different meanings in RAG?

  #18060 closed Mar 8, 2025

• [Feature Request]: Gemini 2.0 from the new GenAI SDK

  #17935 closed Mar 8, 2025

• [Bug]: Timedata Type Metadata Failing on GFS (SimpleDirectoryReader)

  #17907 closed Mar 7, 2025

• [Feature Request]: support stop sequences with `Gemini` class

  #16516 closed Mar 7, 2025

• [Feature Request]: Workflow input/output type annotation

  #16334 closed Mar 6, 2025

• [Documentation]: Tools cannot put the output of one tool into another in a single turn

  #18035 closed Mar 6, 2025

• [Question]: How to deal with JSON with image location

  #15911 closed Mar 6, 2025

26 Issues opened by 25 people

• [Question]: How to use local deployed embeding and chat model

  #18115 opened Mar 13, 2025

• [Bug]: AssertionError for AzureOpenAIMultiModal while using VectorStoreIndex as query engine or as chat engine

  #18111 opened Mar 12, 2025

• [Bug]: When context is used in workflow, there is a memory leak

  #18107 opened Mar 12, 2025

• [Question]: Streaming Response printed all at once

  #18104 opened Mar 12, 2025

• [Feature Request]: llama-index-llms-huggingface-api to llama-index-llms-huggingface-inference-api

  #18101 opened Mar 11, 2025

• [Question]: When using Ollama as LLMs and use function calling, the `astriam_chat` function will not generate stream output

  #18099 opened Mar 11, 2025

• [Bug]: GitHubRepositoryIssuesReader does not follow github rediction

  #18098 opened Mar 11, 2025

• [Bug]: Inconsistent Streaming Response Handling between OpenAI and Gemini LLM Classes

  #18096 opened Mar 11, 2025

• [Question]: Stuck with not getting appropriate answers.

  #18095 opened Mar 11, 2025

• [Question]: ReAct Agent and stream_chat compatible?

  #18094 opened Mar 11, 2025

• QueryFusionRetriever blocks when using use_async

  #18090 opened Mar 11, 2025

• [Bug]: elasticsearch.BadRequestError: BadRequestError(400, 'parsing_exception', 'Unknown key for a START_OBJECT in [retriever].')

  #18089 opened Mar 11, 2025

• [Bug]: Cannot install `llama-index-storage-chat-store-postgres` on Python 3.13 due to outdated `asyncpg` dependency

  #18085 opened Mar 10, 2025

• [Feature Request]: Integrating TMDB and OMDB API with LlamaIndex

  #18071 opened Mar 9, 2025

• [Bug]: Multi-agent Wokflow context serialization issue

  #18069 opened Mar 9, 2025

• [Feature Request]: Implement astream_chat into Vertex

  #18068 opened Mar 9, 2025

• [Bug]: KeyError: 'title' in get_function_tool when using structured_predict with Pydantic models containing Forward References

  #18067 opened Mar 9, 2025

• [Bug]: Asyncio task created without holding a reference to it in Workflow

  #18058 opened Mar 8, 2025

• [Bug]: BaseToolSpec.to_tool_list doesn't work with Workflow Context args

  #18057 opened Mar 8, 2025

• [Feature Request]: Allow Agent to use BashTools

  #18056 opened Mar 8, 2025

• [Bug]: Template duplication in the nodes

  #18053 opened Mar 7, 2025

• [Bug]: Detected nested async when trying to run it with FastAPI

  #18049 opened Mar 7, 2025

• Add "api_version": self.api_version to llms.openai

  #18048 opened Mar 7, 2025

• [Bug]: GROQ with LLAMAINDEX not working on AWS lambda

  #18036 opened Mar 6, 2025

• [Feature Request]: Paragraph Splitter in Node Parsers

  #18033 opened Mar 6, 2025

• [Bug]: AzureOpenAI not recognizing as MultiModal

  #18030 opened Mar 6, 2025

17 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.

• Feature: Added a reader using the multimodal capability of Gemini

  #18020 commented on Mar 12, 2025 • 9 new comments

• open links in new tabs

  #18026 commented on Mar 6, 2025 • 0 new comments

• REBEL - new node postprocessor

  #17590 commented on Mar 12, 2025 • 0 new comments

• add default model and test cases

  #17127 commented on Mar 7, 2025 • 0 new comments

• [Bug]: PydanticUserError: The `__modify_schema__` method is not supported in Pydantic v2. Use `__get_pydantic_json_schema__` instead in class `SecretStr`.

  #16540 commented on Mar 12, 2025 • 0 new comments

• [Bug]: "Multi-Modal LLM using DashScope qwen-vl model for image reasoning" example code raised a ValidationError by pydantic

  #17291 commented on Mar 12, 2025 • 0 new comments

• [Question]: How to stream intermediate steps/tasks for a ReAct Agent?

  #15952 commented on Mar 12, 2025 • 0 new comments

• [Bug]: Whatsappchat loader error because of polar dataframe

  #17794 commented on Mar 12, 2025 • 0 new comments

• [Bug]: BM25Retriever cannot work on chinese

  #13866 commented on Mar 12, 2025 • 0 new comments

• [Bug]: IngestionPipeline fails to serialize transformations properly

  #17532 commented on Mar 11, 2025 • 0 new comments

• [Question]: Orchestrate multiple workflows.

  #15590 commented on Mar 11, 2025 • 0 new comments

• [Bug]: Structured Outputs sometimes return string instead of Pydantic Model

  #16604 commented on Mar 11, 2025 • 0 new comments

• [Bug]: `SimpleDocumentStore.docs` should return documents, but returns nodes?

  #15489 commented on Mar 10, 2025 • 0 new comments

• [Feature Request]: Add dimensions parameter to AzureOpenAIEmbedding

  #17623 commented on Mar 8, 2025 • 0 new comments

• [Question]: VectorStoreIndex not adding embeddings inside the Qdrant DB

  #16981 commented on Mar 8, 2025 • 0 new comments

• [Bug]: VertexAI Integration with ReAct Agent doesn't support Gemini 2.0

  #17932 commented on Mar 6, 2025 • 0 new comments

• [Bug]: Query failure with Document Summary Index using Azure AI Search and Cosmos DB

  #17304 commented on Mar 6, 2025 • 0 new comments