1 Release published by 1 person

• v0.12.2

  published Nov 26, 2024

31 Pull requests merged by 19 people

• feat: improve ImageBlock

  #17111 merged Dec 2, 2024

• Allow install llama-index-packs-code-hierachy in python 3.12 env

  #17104 merged Dec 2, 2024

• Make Weaviate Vector Store integration support nested metadata filtering

  #17107 merged Dec 2, 2024

• Remove forgotten print in ChatMemoryBuffer

  #17114 merged Nov 30, 2024

• Create tables for pgvector regardless of schema status

  #17100 merged Nov 29, 2024

• [FIX] Move JSONalyzeQueryEngine to experimental

  #17110 merged Nov 29, 2024

• fix: Broken links in documentation

  #17097 merged Nov 29, 2024

• fix: use prepared statement in DuckDBRetriever

  #17092 merged Nov 29, 2024

• Bump notebook from 7.0.7 to 7.2.2 in /llama-index-core

  #17093 merged Nov 29, 2024

• chore: fix httpx_client typo in LlamaCloudRetriever

  #17101 merged Nov 29, 2024

• Update upstage solar pro model tokenizer

  #17081 merged Nov 27, 2024

• fix/openai-embbeding-retry

  #17072 merged Nov 27, 2024

• fix/anyscale-embbeding-retry

  #17073 merged Nov 27, 2024

• [DOCS] Add Argilla to integration docs

  #17080 merged Nov 27, 2024

• Clean up samba docs/classes

  #17087 merged Nov 27, 2024

• fix: wrong project id variable in LlamaCloudRetriever

  #17086 merged Nov 27, 2024

• [Integrations]: Bump version for Kùzu

  #17078 merged Nov 27, 2024

• Remove unused store_checkpoints var in tests and example notebook

  #17083 merged Nov 27, 2024

• ci: manually install pants

  #17079 merged Nov 26, 2024

• v0.12.2

  #17076 merged Nov 26, 2024

• feat: Make ChatMessage use ContentBlocks for data storage

  #17039 merged Nov 26, 2024

• remove pyarrow from vertex deps

  #16997 merged Nov 26, 2024

• Add stream_chat and conditionally set AutoModelClass to MllamaForConditionalGeneration

  #17031 merged Nov 26, 2024

• [DOCS] remove Argilla legacy callback and add to instrumentation

  #17067 merged Nov 26, 2024

• [Feature] Checkpointing with Workflows

  #17006 merged Nov 26, 2024

• Fix base component tests

  #17062 merged Nov 26, 2024

• check if schema exists before doing rest of init during pgvector table creation

  #17063 merged Nov 26, 2024

• mark code splitter tests as optional

  #17060 merged Nov 26, 2024

• Fix box api - mypy breakage

  #17061 merged Nov 25, 2024

• - chore: remove example notebook on deprecated pinecone reader

  #17042 merged Nov 25, 2024

• Fix: update perplexity models, add unit tests and minor fixes

  #17045 merged Nov 25, 2024

9 Pull requests opened by 7 people

• chore: standardize FileSystemReaderMixin interface + readers

  #17069 opened Nov 26, 2024

• feat: add get_file_name and get file_path to filesystem readers

  #17075 opened Nov 26, 2024

• [WIP] Update Pgvector Test Coverage

  #17112 opened Nov 29, 2024

• add support for voyageai >=0.3.0

  #17120 opened Dec 2, 2024

• Make safe_pickle_load function.

  #17121 opened Dec 2, 2024

• fix(metrics): fixed NDCG calculation and added comprehensive test cases

  #17126 opened Dec 2, 2024

• add default model and test cases

  #17127 opened Dec 2, 2024

• Add support for IS_EMPTY metadata filters to Weaviate Vector Store integration

  #17128 opened Dec 2, 2024

• Make Weaviate Vector Store integration work with complex properties

  #17129 opened Dec 2, 2024

26 Issues closed by 6 people

• [Feature Request]: Superposition Prompting

  #15644 closed Dec 2, 2024

• [Question]: How to solve the problem of garbled characters caused by pictures in the read doc

  #15636 closed Dec 2, 2024

• [Feature Request]: Metadata filter doesn't support array as filter value and IN operator for filter searching

  #15484 closed Dec 2, 2024

• [Feature Request]: All Conectors/All Integrations

  #15627 closed Dec 1, 2024

• [Bug]: AsyncWebPageReader states that pdf-urls or urls redirecting to pdfs are not valid

  #15625 closed Dec 1, 2024

• [Question]: how to exclude metadata

  #17118 closed Dec 1, 2024

• [Question]: how to remove metadata ？

  #15582 closed Nov 30, 2024

• [Question]: UndefinedTable Error When Using PGVectorStore with LlamaIndex

  #17094 closed Nov 29, 2024

• Move SQLLite package to experimental

  #17109 closed Nov 29, 2024

• How to reuse embeddings when created in elastic search with denseX.

  #15589 closed Nov 29, 2024

• [Bug]: WolframAlphaToolSpec broken

  #15578 closed Nov 29, 2024

• Add a section in our docs for how to debug workflows.

  #16837 closed Nov 29, 2024

• [Bug]: TypeError while importing AzureOpenAI from llama_index.llms.azure_openai

  #17089 closed Nov 28, 2024

• [Question]: How to use WikipediaToolSpec with local Ollama

  #17096 closed Nov 28, 2024

• [Question]: Why package `spacy` is not a dependency of `llama-index-core`?

  #17099 closed Nov 28, 2024

• [Question]: cannot import name 'VectorStoreIndex' from 'llama_index.core

  #17091 closed Nov 28, 2024

• [Bug]: TypeError: AsyncClient.chat() got an unexpected keyword argument 'tools'

  #15319 closed Nov 27, 2024

• [Bug]: OpenAIEmbedding not respecting the given max_retries

  #17071 closed Nov 27, 2024

• [Documentation]: Sambanova Integration Issues: Package Names Mismatch and Import Errors

  #17084 closed Nov 27, 2024

• [Question]: Streaming Response printed all at once instead of Token by Token

  #17024 closed Nov 27, 2024

• [Bug]: not able to retrieve keyword based search

  #17074 closed Nov 27, 2024

• [Question]: llama-index-vector-stores-chroma version for llama-index 0.12.0

  #17068 closed Nov 26, 2024

• [Question]: Multi-doc RAG using Weaviate vector store

  #15530 closed Nov 26, 2024

• [Feature Request]: Implement tool_call_parser on ContextRetrieverOpenAIAgent.from_tools_and_retriever

  #15514 closed Nov 26, 2024

• [Question]: In pgvector store how to add new fields to the table it creates and define relationship among my existing table

  #17022 closed Nov 26, 2024

• [Bug]: Perplexity 400 Client Error: Bad Request for url

  #17044 closed Nov 25, 2024

25 Issues opened by 22 people

• [Bug]: Adding RedisCache to an ingestion pipeline makes it unparallelizable

  #17125 opened Dec 2, 2024

• [Bug]: Authentication Error when Using PropertyGraphIndex with LlamaIndex

  #17124 opened Dec 2, 2024

• [Bug]: Embedding creation with DenseX Retrieval pack

  #17123 opened Dec 2, 2024

• [Question]: Get Node with score with astreamchat

  #17122 opened Dec 2, 2024

• [Question]:VectorStoreIndex and callback_manager- "float division by zero" Error

  #17119 opened Dec 1, 2024

• [Bug]: AssertionError 2024-12-01 12:51:18.470 Examining the path of torch.classes raised: Tried to instantiate class '__path__._path', but it does not exist! Ensure that it is registered via torch::class_

  #17117 opened Dec 1, 2024

• [Question]: Can I use the same gpt-4o-mini model for the whole evaluation process?

  #17116 opened Nov 30, 2024

• [Bug]: New Anthropic tokenizer (PR #16909) breaks the Tokenizer protocol

  #17115 opened Nov 30, 2024

• [Question]: astream_chat is not working whereas achat is working

  #17113 opened Nov 29, 2024

• Move stale projects over to stale branch

  #17108 opened Nov 29, 2024

• [Feature Request]: Metadata Filtering In Query component

  #17106 opened Nov 29, 2024

• [Bug]: MultiModalLLMCompletionProgram (Ollama:Llava) raises AttributeError: 'GenerateResponse' object has no attribute 'items'

  #17105 opened Nov 29, 2024

• [Question]: Document Rag setup using llama index

  #17103 opened Nov 29, 2024

• [Question]: Develop a Multi-Agent RAG System based on Graphs

  #17102 opened Nov 29, 2024

• [Question]: How to set max size of the chunks when using SemanticSplitter?

  #17098 opened Nov 28, 2024

• [Question]: Specifying logit_bias for OpenAI completion APIs

  #17095 opened Nov 28, 2024

• [Question]: How to create Custom Parser for attached data sample ?

  #17090 opened Nov 28, 2024

• [Question]: Sharepoint reader with access token

  #17088 opened Nov 27, 2024

• [Question]: Database Connection Pool Not Effective When Using as_retriever with VectorStoreIndex for Concurrent Data Retrieval

  #17085 opened Nov 27, 2024

• [Bug]: Using Groq integration for structured output tries to calls tools

  #17082 opened Nov 27, 2024

• [Bug]: Calculation of ideal DCG in NDCG is incorrect

  #17070 opened Nov 26, 2024

• [Question]: multiple worrying inconsistencies with openai apis and docs

  #17066 opened Nov 26, 2024

• [Bug]: PropertyGraphIndex (Neo4J Aura): PydanticSerializationError: Error calling function `<lambda>`: AttributeError: 'str' object has no attribute 'value'

  #17065 opened Nov 26, 2024

• [Question]: video summarizer using llama index ?

  #17064 opened Nov 26, 2024

• [Question]: llamaindex filter

  #17059 opened Nov 25, 2024

15 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.

• [Question]: Video Rag documentation

  #17055 commented on Nov 26, 2024 • 0 new comments

• [Feature Request]: Add Bedrock multimodal LLM integration

  #13507 commented on Nov 26, 2024 • 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 Nov 26, 2024 • 0 new comments

• [Feature Request]: Specify inference parameters for Huggingface models using HuggingfaceInferenceAPI

  #15659 commented on Nov 26, 2024 • 0 new comments

• [Bug]: Workflow.run and Workflow.run_step return str. They should return Any

  #15624 commented on Nov 26, 2024 • 0 new comments

• [Bug]: Unable to change MilvusVectorStore uri parameter to a local Milvus path

  #17046 commented on Nov 27, 2024 • 0 new comments

• [Question]: What are versions to be used for an advanced RAG using chromadb,vectorindex llamaindex3, 8b: 8 Billion parameters

  #17000 commented on Nov 28, 2024 • 0 new comments

• use MarkdownElementNodeParser independently

  #16707 commented on Nov 28, 2024 • 0 new comments

• [Feature Request]: Microsoft Exchange Online Reader

  #15751 commented on Nov 29, 2024 • 0 new comments

• Graphrag cookbook httpx timeout error

  #15740 commented on Nov 29, 2024 • 0 new comments

• [Question]: how to insert doc into chrom or bm25 online?

  #15755 commented on Nov 30, 2024 • 0 new comments

• [Question]: What happend in this error

  #16718 commented on Dec 2, 2024 • 0 new comments

• [Feature Request]: Hybrid Search with DatabricksVectorSearch()

  #15738 commented on Dec 2, 2024 • 0 new comments

• [Feature Request]: Implement DIEM as a measure for calculating similar vectors for vector store indices

  #15742 commented on Dec 2, 2024 • 0 new comments

• [Bug]: Stepwise slightly broken for workflows

  #15674 commented on Dec 2, 2024 • 0 new comments