llama-index

llama-index-core [0.10.42]

• Allow proper setting of the vector store in property graph index (#13816)
• fix imports in langchain bridge (#13871)

llama-index-graph-stores-nebula [0.2.0]

• NebulaGraph support for PropertyGraphStore (#13816)

llama-index-llms-langchain [0.1.5]

• fix fireworks imports in langchain llm (#13871)

llama-index-llms-openllm [0.1.5]

• feat(openllm): 0.5 sdk integrations update (#13848)

llama-index-llms-premai [0.1.5]

• Update SDK compatibility (#13836)

llama-index-readers-google [0.2.6]

• Fixed a bug with tokens causing an infinite loop in GoogleDriveReader (#13863)

llama-index-core [0.10.41]

• pass embeddings from index to property graph retriever (#13843)
• protect instrumentation event/span handlers from each other (#13823)
• add missing events for completion streaming (#13824)
• missing callback_manager.on_event_end when there is exception (#13825)

llama-index-llms-gemini [0.1.10]

• use model kwarg for model name for gemini (#13791)

llama-index-llms-mistralai [0.1.15]

• Add mistral code model (#13807)
• update mistral codestral with fill in middle endpoint (#13810)

llama-index-llms-openllm [0.1.5]

• 0.5 integrations update (#13848)

llama-index-llms-vertex [0.1.8]

• Safety setting for Pydantic Error for Vertex Integration (#13817)

llama-index-readers-smart-pdf-loader [0.1.5]

• handle path objects in smart pdf reader (#13847)

llama-index-core [0.10.40]

• Added PropertyGraphIndex and other supporting abstractions. See the full guide for more details (#13747)
• Updated AutoPrevNextNodePostprocessor to allow passing in response mode and LLM (#13771)
• fix type handling with return direct (#13776)
• Correct the method name to _aget_retrieved_ids_and_texts in retrievval evaluator (#13765)
• fix: QueryTransformComponent incorrect call self._query_transform (#13756)
• implement more filters for SimpleVectorStoreIndex (#13365)

llama-index-embeddings-bedrock [0.2.0]

• Added support for Bedrock Titan Embeddings v2 (#13580)

llama-index-embeddings-oci-genai [0.1.0]

• add Oracle Cloud Infrastructure (OCI) Generative AI (#13631)

llama-index-embeddings-huggingface [0.2.1]

• Expose "safe_serialization" parameter from AutoModel (#11939)

llama-index-graph-stores-neo4j [0.2.0]

• Added Neo4jPGStore for property graph support (#13747)

llama-index-indices-managed-dashscope [0.1.1]

• Added dashscope managed index (#13378)

llama-index-llms-oci-genai [0.1.0]

• add Oracle Cloud Infrastructure (OCI) Generative AI (#13631)

llama-index-readers-feishu-wiki [0.1.1]

• fix undefined variable (#13768)

llama-index-packs-secgpt [0.1.0]

• SecGPT - LlamaIndex Integration #13127

llama-index-vector-stores-hologres [0.1.0]

• Add Hologres vector db (#13619)

llama-index-vector-stores-milvus [0.1.16]

• Remove FlagEmbedding as Milvus's dependency (#13767) Unify the collection construction regardless of the value of enable_sparse (#13773)

llama-index-vector-stores-opensearch [0.1.9]

• refactor to put helper methods inside class definition (#13749)

llama-index-agent-introspective [0.1.0]

• Add CRITIC and reflection agent integrations (#13108)

llama-index-core [0.10.35]

• fix from_defaults() erasing summary memory buffer history (#13325)
• use existing async event loop instead of asyncio.run() in core (#13309)
• fix async streaming from query engine in condense question chat engine (#13306)
• Handle ValueError in extract_table_summaries in element node parsers (#13318)
• Handle llm properly for QASummaryQueryEngineBuilder and RouterQueryEngine (#13281)
• expand instrumentation payloads (#13302)
• Fix Bug in sql join statement missing schema (#13277)

llama-index-embeddings-jinaai [0.1.5]

• add encoding_type parameters in JinaEmbedding class (#13172)
• fix encoding type access in JinaEmbeddings (#13315)

llama-index-embeddings-nvidia [0.1.0]

• add nvidia nim embeddings support (#13177)

llama-index-llms-mistralai [0.1.12]

• Fix async issue when streaming with Mistral AI (#13292)

llama-index-llms-nvidia [0.1.0]

• add nvidia nim llm support (#13176)

llama-index-postprocessor-nvidia-rerank [0.1.0]

• add nvidia nim rerank support (#13178)

llama-index-readers-file [0.1.21]

• Update MarkdownReader to parse text before first header (#13327)

llama-index-readers-web [0.1.13]

• feat: Spider Web Loader (#13200)

llama-index-vector-stores-vespa [0.1.0]

• Add VectorStore integration for Vespa (#13213)

llama-index-vector-stores-vertexaivectorsearch [0.1.0]

• Add support for Vertex AI Vector Search as Vector Store (#13186)

llama-index-core [0.10.34]

• remove error ignoring during chat engine streaming (#13160)
• add structured planning agent (#13149)
• update base class for planner agent (#13228)
• Fix: Error when parse file using SimpleFileNodeParser and file's extension doesn't in FILE_NODE_PARSERS (#13156)
• add matching source_node.node_id verification to node parsers (#13109)
• Retrieval Metrics: Updating HitRate and MRR for Evaluation@K documents retrieved. Also adding RR as a separate metric (#12997)
• Add chat summary memory buffer (#13155)

llama-index-indices-managed-zilliz [0.1.3]

• ZillizCloudPipelineIndex accepts flexible params to create pipelines (#10134, #10112)

llama-index-llms-huggingface [0.1.7]

• Add tool usage support with text-generation-inference integration from Hugging Face (#12471)

llama-index-llms-maritalk [0.2.0]

• Add streaming for maritalk (#13207)

llama-index-llms-mistral-rs [0.1.0]

• Integrate mistral.rs LLM (#13105)

llama-index-llms-mymagic [0.1.7]

• mymagicai api update (#13148)

llama-index-llms-nvidia-triton [0.1.5]

• Streaming Support for Nvidia's Triton Integration (#13135)

llama-index-llms-ollama [0.1.3]

• added async support to ollama llms (#13150)

llama-index-readers-microsoft-sharepoint [0.2.2]

• Exclude access control metadata keys from LLMs and embeddings - SharePoint Reader (#13184)

llama-index-readers-web [0.1.11]

• feat: Browserbase Web Reader (#12877)

llama-index-readers-youtube-metadata [0.1.0]

• Added YouTube Metadata Reader (#12975)

llama-index-storage-kvstore-redis [0.1.4]

• fix redis kvstore key that was in bytes (#13201)

llama-index-vector-stores-azureaisearch [0.1.5]

• Respect filter condition for Azure AI Search (#13215)

llama-index-vector-stores-chroma [0.1.7]

• small bump for new chroma client version (#13158)

llama-index-vector-stores-firestore [0.1.0]

• Adding Firestore Vector Store (#12048)

llama-index-vector-stores-kdbai [0.1.5]

• small fix to returned IDs after add() (#12515)

llama-index-vector-stores-milvus [0.1.11]

• Add hybrid retrieval mode to MilvusVectorStore (#13122)

llama-index-vector-stores-postgres [0.1.7]

• parameterize queries in pgvector store (#13199)

llama-index-core [0.10.31]

• fix async streaming response from query engine (#12953)
• enforce uuid in element node parsers (#12951)
• add function calling LLM program (#12980)
• make the PydanticSingleSelector work with async api (#12964)
• fix query pipeline's arun_with_intermediates (#13002)

llama-index-agent-coa [0.1.0]

• Add COA Agent integration (#13043)

llama-index-agent-lats [0.1.0]

• Official LATs agent integration (#13031)

llama-index-agent-llm-compiler [0.1.0]

• Add LLMCompiler Agent Integration (#13044)

llama-index-llms-anthropic [0.1.10]

• Add the ability to pass custom headers to Anthropic LLM requests (#12819)

llama-index-llms-bedrock [0.1.7]

• Adding claude 3 opus to BedRock integration (#13033)

llama-index-llms-fireworks [0.1.5]

• Add new Llama 3 and Mixtral 8x22b model into Llama Index for Fireworks (#12970)

llama-index-llms-openai [0.1.16]

• Fix AsyncOpenAI "RuntimeError: Event loop is closed bug" when instances of AsyncOpenAI are rapidly created & destroyed (#12946)
• Don't retry on all OpenAI APIStatusError exceptions - just InternalServerError (#12947)

llama-index-llms-watsonx [0.1.7]

• Updated IBM watsonx foundation models (#12973)

llama-index-packs-code-hierarchy [0.1.6]

• Return the parent node if the query node is not present (#12983)
• fixed bug when function is defined twice (#12941)

llama-index-program-openai [0.1.6]

• dding support for streaming partial instances of Pydantic output class in OpenAIPydanticProgram (#13021)

llama-index-readers-openapi [0.1.0]

• add reader for openapi files (#12998)

llama-index-readers-slack [0.1.4]

• Avoid infinite loop when not handled exception is raised (#12963)

llama-index-readers-web [0.1.10]

• Improve whole site reader to remove duplicate links (#12977)

llama-index-retrievers-bedrock [0.1.1]

• Fix Bedrock KB retriever to use query bundle (#12910)

llama-index-vector-stores-awsdocdb [0.1.0]

• Integrating AWS DocumentDB as a vector storage method (#12217)

llama-index-vector-stores-databricks [0.1.2]

• Fix databricks vector search metadata (#12999)

llama-index-vector-stores-neo4j [0.1.4]

• Neo4j metadata filtering support (#12923)

llama-index-vector-stores-pinecone [0.1.5]

• Fix error querying PineconeVectorStore using sparse query mode (#12967)

llama-index-vector-stores-qdrant [0.2.5]

• Many fixes for async and checking if collection exists (#12916)

llama-index-vector-stores-weaviate [0.1.5]

• Adds the index deletion functionality to the WeviateVectoreStore (#12993)

llama-index-core [0.10.30]

• Add intermediate outputs to QueryPipeline (#12683)
• Fix show progress causing results to be out of order (#12897)
• add OR filter condition support to simple vector store (#12823)
• improved custom agent init (#12824)
• fix pipeline load without docstore (#12808)
• Use async _aprocess_actions in _arun_step_stream (#12846)
• provide the exception to the StreamChatErrorEvent (#12879)
• fix bug in load and search tool spec (#12902)

llama-index-embeddings-azure-opena [0.1.7]

• Expose azure_ad_token_provider argument to support token expiration (#12818)

llama-index-embeddings-cohere [0.1.8]

• Add httpx_async_client option (#12896)

llama-index-embeddings-ipex-llm [0.1.0]

• add ipex-llm embedding integration (#12740)

llama-index-embeddings-octoai [0.1.0]

• add octoai embeddings (#12857)

llama-index-llms-azure-openai [0.1.6]

• Expose azure_ad_token_provider argument to support token expiration (#12818)

llama-index-llms-ipex-llm [0.1.2]

• add support for loading "low-bit format" model to IpexLLM integration (#12785)

llama-index-llms-mistralai [0.1.11]

• support open-mixtral-8x22b (#12894)

llama-index-packs-agents-lats [0.1.0]

• added LATS agent pack (#12735)

llama-index-readers-smart-pdf-loader [0.1.4]

• Use passed in metadata for documents (#12844)

llama-index-readers-web [0.1.9]

• added Firecrawl Web Loader (#12825)

llama-index-vector-stores-milvus [0.1.10]

• use batch insertions into Milvus vector store (#12837)

llama-index-vector-stores-vearch [0.1.0]

• add vearch to vector stores (#10972)

llama-index-core [0.10.29]

• BREAKING Moved PandasQueryEngine and PandasInstruction parser to llama-index-experimental (#12419)
  • new install: pip install -U llama-index-experimental
  • new import: from llama_index.experimental.query_engine import PandasQueryEngine
• Fixed some core dependencies to make python3.12 work nicely (#12762)
• update async utils run_jobs() to include tqdm description (#12812)
• Refactor kvdocstore delete methods (#12681)

llama-index-llms-bedrock [0.1.6]

• Support for Mistral Large from Bedrock (#12804)

llama-index-llms-openvino [0.1.0]

• Added OpenVino LLMs (#12639)

llama-index-llms-predibase [0.1.4]

• Update LlamaIndex-Predibase Integration to latest API (#12736)
• Enable choice of either Predibase-hosted or HuggingFace-hosted fine-tuned adapters in LlamaIndex-Predibase integration (#12789)

llama-index-output-parsers-guardrails [0.1.3]

• Modernize GuardrailsOutputParser (#12676)

llama-index-packs-agents-coa [0.1.0]

• Chain-of-Abstraction Agent Pack (#12757)

llama-index-packs-code-hierarchy [0.1.3]

• Fixed issue with chunking multi-byte characters (#12715)

llama-index-packs-raft-dataset [0.1.4]

• Fix bug in raft dataset generator - multiple system prompts (#12751)

llama-index-postprocessor-openvino-rerank [0.1.2]

• Add openvino rerank support (#12688)

llama-index-readers-file [0.1.18]

• convert to Path in docx reader if input path str (#12807)
• make pip check work for optional pdf packages (#12758)

llama-index-readers-s3 [0.1.7]

• wrong doc id when using default s3 endpoint in S3Reader (#12803)

llama-index-retrievers-bedrock [0.1.0]

• Add Amazon Bedrock knowledge base integration as retriever (#12737)

llama-index-retrievers-mongodb-atlas-bm25-retriever [0.1.3]

• Add mongodb atlas bm25 retriever (#12519)

llama-index-storage-chat-store-redis [0.1.3]

• fix message serialization in redis chat store (#12802)

llama-index-vector-stores-astra-db [0.1.6]

• Relax dependency version to accept astrapy 1.* (#12792)

llama-index-vector-stores-couchbase [0.1.0]

• Add support for Couchbase as a Vector Store (#12680)

llama-index-vector-stores-elasticsearch [0.1.7]

• Fix elasticsearch hybrid rrf window_size (#12695)

llama-index-vector-stores-milvus [0.1.8]

• Added support to retrieve metadata fields from milvus (#12626)

llama-index-vector-stores-redis [0.2.0]

• Modernize redis vector store, use redisvl (#12386)

llama-index-vector-stores-qdrant [0.2.0]

• refactor: Switch default Qdrant sparse encoder (#12512)

llama-index-core [0.10.28]

• Support indented code block fences in markdown node parser (#12393)
• Pass in output parser to guideline evaluator (#12646)
• Added example of query pipeline + memory (#12654)
• Add missing node postprocessor in CondensePlusContextChatEngine async mode (#12663)
• Added return_direct option to tools /tool metadata (#12587)
• Add retry for batch eval runner (#12647)
• Thread-safe instrumentation (#12638)
• Coroutine-safe instrumentation Spans #12589
• Add in-memory loading for non-default filesystems in PDFReader (#12659)
• Remove redundant tokenizer call in sentence splitter (#12655)
• Add SynthesizeComponent import to shortcut imports (#12655)
• Improved truncation in SimpleSummarize (#12655)
• adding err handling in eval_utils default_parser for correctness (#12624)
• Add async_postprocess_nodes at RankGPT Postprocessor Nodes (#12620)
• Fix MarkdownNodeParser ref_doc_id (#12615)

llama-index-embeddings-openvino [0.1.5]

• Added initial support for openvino embeddings (#12643)

llama-index-llms-anthropic [0.1.9]

• add anthropic tool calling (#12591)

llama-index-llms-ipex-llm [0.1.1]

• add ipex-llm integration (#12322)
• add more data types support to ipex-llm llm integration (#12635)

llama-index-llms-openllm [0.1.4]

• Proper PrivateAttr usage in OpenLLM (#12655)

llama-index-multi-modal-llms-anthropic [0.1.4]

• Bumped anthropic dep version (#12655)

llama-index-multi-modal-llms-gemini [0.1.5]

• bump generativeai dep (#12645)

llama-index-packs-dense-x-retrieval [0.1.4]

• Add streaming support for DenseXRetrievalPack (#12607)

llama-index-readers-mongodb [0.1.4]

• Improve efficiency of MongoDB reader (#12664)

llama-index-readers-wikipedia [0.1.4]

• Added multilingual support for the Wikipedia reader (#12616)

llama-index-storage-index-store-elasticsearch [0.1.3]

• remove invalid chars from default collection name (#12672)

llama-index-vector-stores-milvus [0.1.8]

• Added support to retrieve metadata fields from milvus (#12626)
• Bug fix - Similarity metric is always IP for MilvusVectorStore (#12611)

llama-index-agent-openai [0.2.2]

• Update imports for message thread typing (#12437)

llama-index-core [0.10.27]

• Fix for pydantic query engine outputs being blank (#12469)
• Add span_id attribute to Events (instrumentation) (#12417)
• Fix RedisDocstore node retrieval from docs property (#12324)
• Add node-postprocessors to retriever_tool (#12415)
• FLAREInstructQueryEngine : delegating retriever api if the query engine supports it (#12503)
• Make chat message to dict safer (#12526)
• fix check in batch eval runner for multi-kwargs (#12563)
• Fixes agent_react_multimodal_step.py bug with partial args (#12566)

llama-index-embeddings-clip [0.1.5]

• Added support to load clip model from local file path (#12577)

llama-index-embeddings-cloudflar-workersai [0.1.0]

• text embedding integration: Cloudflare Workers AI (#12446)

llama-index-embeddings-voyageai [0.1.4]

• Fix pydantic issue in class definition (#12469)

llama-index-finetuning [0.1.5]

• Small typo fix in QA generation prompt (#12470)

llama-index-graph-stores-falkordb [0.1.3]

• Replace redis driver with FalkorDB driver (#12434)

llama-index-llms-anthropic [0.1.8]

• Add ability to pass custom HTTP headers to Anthropic client (#12558)

llama-index-llms-cohere [0.1.6]

• Add support for Cohere Command R+ model (#12581)

llama-index-llms-databricks [0.1.0]

• Integrations with DataBricks LLM API (#12432)

llama-index-llms-watsonx [0.1.6]

• Updated Watsonx foundation models (#12493)
• Updated base model name on watsonx integration #12491

lama-index-postprocessor-rankllm-rerank [0.1.2]

• Add RankGPT support inside RankLLM (#12475)

llama-index-readers-microsoft-sharepoint [0.1.7]

• Use recursive strategy by default for SharePoint (#12557)

llama-index-readers-web [0.1.8]

• Readability web page reader fix playwright async api bug (#12520)

llama-index-vector-stores-kdbai [0.1.5]

• small to_list fix (#12515)

llama-index-vector-stores-neptune [0.1.0]

• Add support for Neptune Analytics as a Vector Store (#12423)

llama-index-vector-stores-postgres [0.1.5]

• fix(postgres): numeric metadata filters (#12583)

llama-index-cli [0.1.9]

• Removed chroma as a bundled dep to reduce llama-index deps

llama-index-core [0.10.19]

• Introduce retries for rate limits in OpenAI llm class (#11867)
• Added table comments to SQL table schemas in SQLDatabase (#11774)
• Added LogProb type to ChatResponse object (#11795)
• Introduced LabelledSimpleDataset (#11805)
• Fixed insert IndexNode objects with unserializable objects (#11836)
• Fixed stream chat type error when writing response to history in CondenseQuestionChatEngine (#11856)
• Improve post-processing for json query engine (#11862)

llama-index-embeddings-cohere [0.1.4]

• Fixed async kwarg error (#11822)

llama-index-embeddings-dashscope [0.1.2]

• Fixed pydantic import (#11765)

llama-index-graph-stores-neo4j [0.1.3]

• Properly close connection after verifying connectivity (#11821)

llama-index-llms-cohere [0.1.3]

• Add support for new command-r model (#11852)

llama-index-llms-huggingface [0.1.4]

• Fixed streaming decoding with special tokens (#11807)

llama-index-llms-mistralai [0.1.5]

• Added support for latest and open models (#11792)

llama-index-tools-finance [0.1.1]

• Fixed small bug when passing in the API get for stock news (#11772)

llama-index-vector-stores-chroma [0.1.6]

• Slimmed down chroma deps (#11775)

llama-index-vector-stores-lancedb [0.1.3]

• Fixes for deleting (#11825)

llama-index-vector-stores-postgres [0.1.3]

• Support for nested metadata filters (#11778)

[2024-03-06]

New format! Going to try out reporting changes per package.

llama-index-cli [0.1.8]

• Update mappings for upgrade command (#11699)

llama-index-core [0.10.17]

• add relative_score and dist_based_score to QueryFusionRetriever (#11667)
• check for none in async agent queue (#11669)
• allow refine template for BaseSQLTableQueryEngine (#11378)
• update mappings for llama-packs (#11699)
• fixed index error for extacting rel texts in KG index (#11695)
• return proper response types from synthesizer when no nodes (#11701)
• Inherit metadata to summaries in DocumentSummaryIndex (#11671)
• Inherit callback manager in sql query engines (#11662)
• Fixed bug with agent streming not being written to chat history (#11675)
• Fixed a small bug with none deltas when streaming a function call with an agent (#11713)

llama-index-multi-modal-llms-anthropic [0.1.2]

• Added support for new multi-modal models haiku and sonnet (#11656)

llama-index-packs-finchat [0.1.0]

• Added a new llama-pack for hierarchical agents + finance chat (#11387)

llama-index-readers-file [0.1.8]

• Added support for checking if NLTK files are already downloaded (#11676)

llama-index-readers-json [0.1.4]

• Use the metadata passed in when creating documents (#11626)

llama-index-vector-stores-astra-db [0.1.4]

• Update wording in warning message (#11702)

llama-index-vector-stores-opensearch [0.1.7]

• Avoid calling nest_asyncio.apply() in code to avoid confusing errors for users (#11707)

llama-index-vector-stores-qdrant [0.1.4]

• Catch RPC errors (#11657)

New Features

• Released llama-index-networks (#11413)
• Jina reranker (#11291)
• Added DuckDuckGo agent search tool (#11386)
• helper functions for chatml (#10272)
• added brave search tool for agents (#11468)
• Added Friendli LLM integration (#11384)
• metadata only queries for chromadb (#11328)

Bug Fixes / Nits

• Fixed inheriting llm callback in synthesizers (#11404)
• Catch delete error in milvus (#11315)
• Fixed pinecone kwargs issue (#11422)
• Supabase metadata filtering fix (#11428)
• api base fix in gemini embeddings (#11393)
• fix elasticsearch vector store await (#11438)
• vllm server cuda fix (#11442)
• fix for passing LLM to context chat engine (#11444)
• set input types for cohere embeddings (#11288)
• default value for azure ad token (#10377)
• added back prompt mixin for react agent (#10610)
• fixed system roles for gemini (#11481)
• fixed mean agg pooling returning numpy float values (#11458)
• improved json path parsing for JSONQueryEngine (#9097)

New Features

• Added a llama-pack for KodaRetriever, for on-the-fly alpha tuning (#11311)
• Added support for mistral-large (#11398)
• Last token pooling mode for huggingface embeddings models like SFR-Embedding-Mistral (#11373)
• Added fsspec support to SimpleDirectoryReader (#11303)

Bug Fixes / Nits

• Fixed an issue with context window + prompt helper (#11379)
• Moved OpenSearch vector store to BasePydanticVectorStore (#11400)
• Fixed function calling in fireworks LLM (#11363)
• Made cohere embedding types more automatic (#11288)
• Improve function calling in react agent (#11280)
• Fixed MockLLM imports (#11376)

New Features

• Added Self-Discover llamapack (#10951)

Bug Fixes / Nits

• Fixed linting in CICD (#10945)
• Fixed using remote graph stores (#10971)
• Added missing LLM kwarg in NoText response synthesizer (#10971)
• Fixed openai import in rankgpt (#10971)
• Fixed resolving model name to string in openai embeddings (#10971)
• Off by one error in sentence window node parser (#10971)

[0.10.6] - 2024-02-17

First, apologies for missing the changelog the last few versions. Trying to figure out the best process with 400+ packages.

At some point, each package will have a dedicated changelog.

But for now, onto the "master" changelog.

New Features

• Added NomicHFEmbedding (#10762)
• Added MinioReader (#10744)

Bug Fixes / Nits

• Various fixes for clickhouse vector store (#10799)
• Fix index name in neo4j vector store (#10749)
• Fixes to sagemaker embeddings (#10778)
• Fixed performance issues when splitting nodes (#10766)
• Fix non-float values in reranker + b25 (#10930)
• OpenAI-agent should be a dep of openai program (#10930)
• Add missing shortcut imports for query pipeline components (#10930)
• Fix NLTK and tiktoken not being bundled properly with core (#10930)
• Add back llama_index.core.__version__ (#10930)

Today we’re excited to launch LlamaIndex v0.10.0. It is by far the biggest update to our Python package to date (see this gargantuan PR), and it takes a massive step towards making LlamaIndex a next-generation, production-ready data framework for your LLM applications.

LlamaIndex v0.10 contains some major updates:

• We have created a llama-index-core package, and split all integrations and templates into separate packages: Hundreds of integrations (LLMs, embeddings, vector stores, data loaders, callbacks, agent tools, and more) are now versioned and packaged as a separate PyPI packages, while preserving namespace imports: for example, you can still usefrom llama_index.llms.openai import OpenAI for a LLM.
• LlamaHub will be the central hub for all integrations: the former llama-hub repo itself is consolidated into the main llama_index repo. Instead of integrations being split between the core library and LlamaHub, every integration will be listed on LlamaHub. We are actively working on updating the site, stay tuned!
• ServiceContext is deprecated: Every LlamaIndex user is familiar with ServiceContext, which over time has become a clunky, unneeded abstraction for managing LLMs, embeddings, chunk sizes, callbacks, and more. As a result we are completely deprecating it; you can now either directly specify arguments or set a default.

Upgrading your codebase to LlamaIndex v0.10 may lead to some breakages, primarily around our integrations/packaging changes, but fortunately we’ve included some scripts to make it as easy as possible to migrate your codebase to use LlamaIndex v0.10.

Full Blog Post

v0.10 Documentation

v0.10 Installation Guide

v0.10 Quickstart

Updated Contribution Guide

Temporary v0.10 Package Registry

v0.10 Migration Guide