1 Release published by 1 person

• v0.12.6

  published Dec 18, 2024

40 Pull requests merged by 24 people

• o1 models

  #17307 merged Dec 18, 2024

• v0.12.6

  #17305 merged Dec 18, 2024

• Update Cerebras examples to use Llama 3.3 70b

  #17301 merged Dec 17, 2024

• Amazon bedrock guardrails

  #17281 merged Dec 17, 2024

• feat: Add tool support to AnthropicMultiModal

  #17302 merged Dec 17, 2024

• Update anthropic type imports for v0.41.0 release

  #17299 merged Dec 17, 2024

• [bug fix] Ensure that StopEvent gets cleared from Context._in_progress["_done"] after a Workflow run

  #17300 merged Dec 17, 2024

• ScrapegraphAI link was not correct and I fix it

  #17292 merged Dec 17, 2024

• fix: add a timeout to langchain callback handler

  #17296 merged Dec 17, 2024

• - add readme to google tools

  #17275 merged Dec 16, 2024

• VoyageAI multimodal embedding, correction

  #17284 merged Dec 16, 2024

• Update Tavily integration documentation

  #17285 merged Dec 16, 2024

• Add text and embedding type to neo4j enhanced schema

  #17289 merged Dec 16, 2024

• Support Confluence cookies

  #17276 merged Dec 16, 2024

• add tool call id/name to mistral chat messages

  #17280 merged Dec 16, 2024

• fix links in RAG understanding

  #17279 merged Dec 16, 2024

• Do not send blank content to anthropic

  #17278 merged Dec 15, 2024

• add content blocks to anthropic

  #17274 merged Dec 15, 2024

• User vs tool

  #17273 merged Dec 15, 2024

• Introducting VoyageAI's new multimodal embeddings model

  #17261 merged Dec 15, 2024

• Add timeout config to neo4j graph

  #17267 merged Dec 15, 2024

• feat: Add NOT filter condition to MetadataFilter and QdrantVectorStore

  #17270 merged Dec 15, 2024

• fix: make ImageDocument derive from Document, backward compatible

  #17259 merged Dec 12, 2024

• Fix: TGI context window

  #17252 merged Dec 11, 2024

• fix openai message dicts for tool calls

  #17254 merged Dec 11, 2024

• Adding llama 3.3-70b as function-calling-capable

  #17253 merged Dec 11, 2024

• Gemini 2.0 support

  #17249 merged Dec 11, 2024

• Update link for the Weaviate Python docs

  #17248 merged Dec 11, 2024

• Handle empty retrieved Pinecone index values

  #17242 merged Dec 11, 2024

• feat: rewrite the example using AzureOpenAI

  #17245 merged Dec 11, 2024

• Add Scrapegraph tool integration

  #17238 merged Dec 11, 2024

• Fix content blocks again

  #17247 merged Dec 11, 2024

• Add ILIKE feature for PGVectorStore and NileVectorStore.

  #17211 merged Dec 11, 2024

• fix: accept already base64-encoded data in ImageBlock

  #17244 merged Dec 11, 2024

• fix: update unstructured dependency pin

  #17246 merged Dec 11, 2024

• fix(metrics): fixed NDCG calculation and updated previous tests

  #17236 merged Dec 11, 2024

• refact: Refactor Document to be natively multimodal

  #17204 merged Dec 11, 2024

• [Issue #17219] MilvusVectorStore Parse "milvus_search_config" out of "vector_store_kwargs"

  #17221 merged Dec 11, 2024

• Add async support to weaviate vector store integration

  #17220 merged Dec 11, 2024

• refactor and optimize milvus code

  #17229 merged Dec 11, 2024

7 Pull requests opened by 7 people

• Doc Error: modify part 2 of advanced prompts

  #17255 opened Dec 12, 2024

• Update base.py

  #17269 opened Dec 13, 2024

• CosmosDB insertion bugfix

  #17290 opened Dec 16, 2024

• fix: allow file extractors to work

  #17294 opened Dec 17, 2024

• feat: add py.typed for OpenAI and Gemini

  #17295 opened Dec 17, 2024

• Update API Reference and Component Guides For Sambanova

  #17306 opened Dec 18, 2024

• VoyageAI Reranker optional API Key

  #17310 opened Dec 18, 2024

18 Issues closed by 7 people

• [Bug]: 'list' object has no attribute 'get_doc_id'

  #17293 closed Dec 17, 2024

• [Question]: Help with RAG and LlamaIndex Response Synthesis

  #17283 closed Dec 16, 2024

• [Question]: How to set system prompts in RAG?

  #17282 closed Dec 16, 2024

• [Bug]: OptimumEmbedding(BaseEmbedding) cannot be selected between CUDA and CPU

  #15908 closed Dec 16, 2024

• [Bug]: Bug in Mistral AI Integration: Missing Tool Call ID in Function Calling

  #17277 closed Dec 16, 2024

• [Bug]: The links in understanding/rag are all broken.

  #17272 closed Dec 16, 2024

• [Feature Request]: NOT filter condition

  #17206 closed Dec 15, 2024

• [Question]: Unable to evaluate RAG Application.

  #15893 closed Dec 14, 2024

• [Question]: I use the faiss vector store, which consists of many nodes (not many documents). how should I get all nodes and their text_embedding from the store_path of faiss vector store

  #15873 closed Dec 13, 2024

• [Feature Request]: update tenacity to 9.0.0 to fix conflict with GraphRAG

  #16876 closed Dec 13, 2024

• [Question]: Dimension issue with fine-tuning using a custom adapter

  #15849 closed Dec 12, 2024

• SPAM CONTACT DEVICE

  #17260 closed Dec 12, 2024

• [Bug]: Cypher error while indexing documents using PropertyGraphIndex

  #17227 closed Dec 12, 2024

• AttributeError: 'ServiceContext' object has no attribute 'add'

  #17256 closed Dec 12, 2024

• [Bug]: Text Generation Inference Context Window Error

  #17251 closed Dec 11, 2024

• [Question]: Inaccurate Responses in RAG System Using LlamaIndex and HuggingFaceLLM

  #15801 closed Dec 11, 2024

• Handling Excel Data with KnowledgeGraphIndex from LlamaIndex

  #15780 closed Dec 11, 2024

• [Feature Request]: MilvusVectorStore - Add ability to specify search_params to query

  #17219 closed Dec 11, 2024

20 Issues opened by 20 people

• [Question]: PrivateGPT only uses cosine similarity with PgVector even with HNSW index added

  #17311 opened Dec 18, 2024

• [Question]: How to pass images into llama index query engine.

  #17309 opened Dec 18, 2024

• [Feature Request]: Specify index_name while creating llama-index MilvusVectorStore instance

  #17308 opened Dec 18, 2024

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

  #17304 opened Dec 17, 2024

• [Question]: SchemaLLMPathExtractor w/ Sonnet Fails for some Schemas

  #17303 opened Dec 17, 2024

• [Bug]: anthropic typing update for v0.41.0 release

  #17298 opened Dec 17, 2024

• [Question]: BAAI/bge-small-en-v1.5 Fails to Load for HuggingFaceEmbedding in LlamaIndex (Stuck Without Progress)

  #17297 opened Dec 17, 2024

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

  #17291 opened Dec 17, 2024

• [Bug]: PropertyGraphIndex Value for metadata features must be one of (str, int, float, None)

  #17288 opened Dec 16, 2024

• [Bug]: Azure CosmosDB NoSQL indexing fails with TypeError: Object of type date is not JSON serializable

  #17287 opened Dec 16, 2024

• [Bug]:

  #17286 opened Dec 16, 2024

• [Feature Request]: OpenSearch Byte Vector Support

  #17271 opened Dec 13, 2024

• [Question]: How can I customize the KnowledgeGraph indexing and querying? How can I load the persisted index?

  #17266 opened Dec 13, 2024

• [Bug]: BedrockConverse updating the system_prompt affects different chat_engines

  #17265 opened Dec 13, 2024

• [Bug]: ValueError: "DashScopeEmbedding" object has no field "_api_key" when initializing DashScopeEmbedding with valid parameters.

  #17264 opened Dec 13, 2024

• [Bug]: ValueError: Can't patch loop of type <class 'uvloop.Loop'>

  #17263 opened Dec 13, 2024

• [Bug]: MetadataFilter value keyword is Optional but it is not assigned as None

  #17262 opened Dec 13, 2024

• [Bug]: Duplicated document object content

  #17258 opened Dec 12, 2024

• [Question]: chunk_size and chunk_overlap are inaccurate

  #17257 opened Dec 12, 2024

• [Bug]: PostgreSQL VectorStore MetadataFilter Lacks FilterOperator.IS_EMPTY Support

  #17250 opened Dec 11, 2024

20 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.

• Vntuananhbui multimodal rag guardrail

  #17177 commented on Dec 17, 2024 • 4 new comments

• [wip] adding workflow tool

  #17237 commented on Dec 11, 2024 • 2 new comments

• Support for multiple indexes in a vector store

  #16972 commented on Dec 13, 2024 • 2 new comments

• Add OCI Data Science Model Deployment Embedding Integration

  #17243 commented on Dec 15, 2024 • 0 new comments

• Add OCI Data Science Model Deployment LLM Integration

  #17241 commented on Dec 15, 2024 • 0 new comments

• add default model and test cases

  #17127 commented on Dec 12, 2024 • 0 new comments

• Add OCI Generative AI tool calling support

  #16888 commented on Dec 17, 2024 • 0 new comments

• [Question]: Implementing Routing Mechanism between two Chat Engines

  #15973 commented on Dec 18, 2024 • 0 new comments

• [Feature Request]: Add external/custom knowledge graphs from Neo4j

  #16055 commented on Dec 17, 2024 • 0 new comments

• [Feature Request]: Llama Index Support for Amazon Bedrock Guardrails

  #17217 commented on Dec 17, 2024 • 0 new comments

• [Feature Request]: Add VoyageAI voyage-multimodal-3 embeddings

  #16985 commented on Dec 17, 2024 • 0 new comments

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

  #17105 commented on Dec 16, 2024 • 0 new comments

• [Feature Request]: Improve ElasticSearch metadata filtering handling

  #14129 commented on Dec 16, 2024 • 0 new comments

• Consolidate MultiModal LLMs with Base LLMs

  #15949 commented on Dec 16, 2024 • 0 new comments

• [Feature Request]: add store to connect searxng as RAG source

  #16020 commented on Dec 14, 2024 • 0 new comments

• [Bug]: Pandas Output: There was an error running the output as Python code. Error message: name 'p' is not defined

  #13800 commented on Dec 14, 2024 • 0 new comments

• [EPIC] MultiModal Refactor

  #15950 commented on Dec 13, 2024 • 0 new comments

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

  #16604 commented on Dec 12, 2024 • 0 new comments

• [Question]: Retrieving Log Probabilities in Streaming Response Using SimpleChatEngine

  #16264 commented on Dec 12, 2024 • 0 new comments

• [Bug]: A PGVector store using a default retriever will not return nodes for large indices

  #17180 commented on Dec 11, 2024 • 0 new comments