◆ AI Agent Frameworks

LlamaIndex

v0.14.24 open-source

LlamaIndex is the leading document agent and OCR platform

source ↗ website ↗ rss ↗ last repo push Aug 19, 2026

from llama_index.llms.minimax import MiniMax

llm = MiniMax()  # defaults to M2.7
response = llm.complete('Summarize the OWASP Top 10 for 2025.')
print(response)
from llama_index.llms.litellm import LiteLLM

llm = LiteLLM(
    model='openai/gpt-4o',
    model_kwargs={'custom_llm_provider': 'azure'}
)
response = llm.complete('List the top cloud misconfigurations.')
print(response)
from llama_index.core.node_parser import SemanticDoubleMergingSplitterNodeParser
from llama_index.embeddings.openai import OpenAIEmbedding

parser = SemanticDoubleMergingSplitterNodeParser(
    embed_model=OpenAIEmbedding(model="text-embedding-3-small")
)
from llama_index.graph_stores.neo4j import Neo4jGraphStore

graph_store = Neo4jGraphStore(
    username="neo4j",
    password="<password>",
    url="bolt://localhost:7687",
    apoc_sample=0.1,
)
from llama_index.tools.mcp import McpToolSpec

tool_spec = McpToolSpec(
    server_url='http://localhost:8080',
    partial_params={'environment': 'production', 'tenant_id': 'acme'}
)
tools = tool_spec.to_tool_list()
from llama_index.ingestion.ray import RayIngestionPipeline

pipeline = RayIngestionPipeline(
    transformations=[...],
    vector_store=my_vector_store,
)
pipeline.run(documents=my_documents)
from llama_index.core.agent.workflow import AgentWorkflow

workflow = AgentWorkflow(
    agents=[...],
    early_stopping_method="generate",
)
from llama_index.embeddings.ollama import OllamaEmbedding

embed_model = OllamaEmbedding(
    model_name="nomic-embed-text",
    keep_alive="10m",
)
embeddings = embed_model.get_text_embedding("Hello, world!")
from llama_index.readers.web import ScrapyWebReader

reader = ScrapyWebReader()
documents = reader.load_data(urls=["https://example.com"])
from llama_index.retrievers.bedrock import AmazonKnowledgeBasesRetriever

retriever = AmazonKnowledgeBasesRetriever(
    knowledge_base_id='<knowledge_base_id>',
    retrieval_config={'vectorSearchConfiguration': {'numberOfResults': 5}},
)
results = await retriever.aretrieve('What is our incident response policy?')
from llama_index.llms.sglang import SGLang

llm = SGLang(model="meta-llama/Llama-3.1-8B-Instruct")
response = llm.complete("Explain prompt injection attacks in one paragraph.")
print(response)
async for event in agent.astream_chat("Explain RSA encryption"):
    if hasattr(event, 'thinking_delta') and event.thinking_delta:
        print("[thinking]", event.thinking_delta)
    if hasattr(event, 'delta') and event.delta:
        print(event.delta, end="", flush=True)
from llama_index.llms.bedrock_converse import BedrockConverse

llm = BedrockConverse(
    model="anthropic.claude-3-5-sonnet-20241022-v2:0",
    system_prompt="You are a security analyst assistant.",
    tool_caching=True,
)
from llama_index.vector_stores.chroma import ChromaVectorStore
from llama_index.core.vector_stores.types import VectorStoreQuery, VectorStoreQueryMode

vector_store = ChromaVectorStore(chroma_collection=collection)
query = VectorStoreQuery(
    query_embedding=embedding,
    similarity_top_k=10,
    mode=VectorStoreQueryMode.MMR,
)
results = vector_store.query(query)
from llama_index.llms.openai import OpenAI

llm = OpenAI(model="gpt-5-chat-latest")
response = llm.complete("Summarize the MITRE ATT&CK framework in three sentences.")
print(response)
from llama_index.readers.s3 import S3Reader

reader = S3Reader(
    bucket="my-bucket",
    client_kwargs={"region_name": "eu-west-1"}
)
documents = reader.load_data()
from llama_index.core.agent.workflow import FunctionAgent

agent = FunctionAgent(
    tools=[my_tool],
    llm=llm,
    system_prompt="You are a helpful assistant."
)
response = await agent.run("What is the capital of France?")
result = agent.run('Summarize the top 5 findings from this report', max_iterations=20)
from llama_index.embeddings.azure_openai import AzureOpenAIEmbedding

embed_model = AzureOpenAIEmbedding(
    model="text-embedding-3-large",
    deployment_name="my-embedding-deployment",
    dimensions=512,
    azure_endpoint="https://<your-resource>.openai.azure.com/",
    api_key="<your-api-key>",
)
from llama_index.core.llms import ChatMessage
from llama_index.core.base.llms.types import CachePoint

messages = [
    ChatMessage(role="user", content=[
        {"type": "text", "text": "You are a helpful assistant with a large knowledge base."},
        CachePoint(),
    ])
]
from llama_index.postprocessor.sbert_rerank import SentenceTransformerRerank

reranker = SentenceTransformerRerank(
    model="cross-encoder/ms-marco-MiniLM-L-6-v2",
    top_n=5,
    cross_encoder_kwargs={"device": "cuda", "max_length": 512},
)
from llama_index.utils.workflow import draw_all_possible_flows

draw_all_possible_flows(MyWorkflow, filename="workflow.html")
from llama_index.tools.mcp import BasicMCPClient

client = BasicMCPClient(
    url="https://my-mcp-server.example.com",
    headers={"Authorization": "Bearer <token>"}
)
from llama_index.core.agent import FunctionAgent

agent = FunctionAgent(
    tools=[...],
    llm=llm,
    allow_parallel_tool_calls=True,
)
from llama_index.core.query_engine import NLSQLTableQueryEngine

query_engine = NLSQLTableQueryEngine(
    sql_database=sql_database,
    row_retriever=row_retriever,
    col_retriever=col_retriever,
    table_retriever=table_retriever,
)
from llama_index.core.workflow import Workflow, step

class MyWorkflow(Workflow):
    @step
    async def my_step(self, ctx, ev):
        return ...
from llama_index.packs.longrag import LongRAGPack

pack = LongRAGPack(documents=documents, llm=llm)
response = pack.run("What are the key findings in this report?")
print(response)
from llama_index.core.retrievers import VectorContextRetriever

retriever = VectorContextRetriever(
    vector_store_index,
    similarity_score=0.75
)
nodes = retriever.retrieve("What is the access control policy?")
from llama_index.vector_stores.pinecone import PineconeVectorStore

vector_store = PineconeVectorStore(pinecone_index=pinecone_index)
vector_store.clear()
from llama_index.embeddings.litellm import LiteLLMEmbedding

embed_model = LiteLLMEmbedding(model="text-embedding-ada-002", api_base="http://localhost:8000")
from llama_index.readers.notion import NotionPageReader

reader = NotionPageReader(integration_token="<token>")
databases = reader.list_databases()
print(databases)
from llama_index.core.vector_stores.types import MetadataFilters, MetadataFilter, FilterCondition

filters = MetadataFilters(
    filters=[
        MetadataFilter(key="source", value="arxiv"),
        MetadataFilter(key="source", value="pubmed"),
    ],
    condition=FilterCondition.OR,
)
results = index.as_retriever(filters=filters).retrieve("transformer models")
from azure.identity import DefaultAzureCredential, get_bearer_token_provider
from llama_index.embeddings.azure_openai import AzureOpenAIEmbedding

token_provider = get_bearer_token_provider(
    DefaultAzureCredential(), "https://cognitiveservices.azure.com/.default"
)
embed_model = AzureOpenAIEmbedding(
    model="text-embedding-ada-002",
    deployment_name="my-deployment",
    azure_endpoint="https://<your-resource>.openai.azure.com/",
    azure_ad_token_provider=token_provider,
)
from llama_index.core.tools import FunctionTool

def lookup_price(ticker: str) -> str:
    return f"${ticker}: 142.00"

price_tool = FunctionTool.from_defaults(
    fn=lookup_price,
    return_direct=True,
)
from llama_index.readers.wikipedia import WikipediaReader

reader = WikipediaReader()
docs = reader.load_data(pages=["Louvre"], lang="fr")
from llama_index.core.retrievers import QueryFusionRetriever

retriever = QueryFusionRetriever(
    retrievers=[retriever_a, retriever_b],
    mode="dist_based_score",
    num_queries=4,
)
nodes = retriever.retrieve("What is the capital of France?")
from llama_index.llms.openai import OpenAI

llm = OpenAI(model="gpt-4")
python gpt_index/tools/migrate_v1_to_v2.py --v1_path tree_v1.json --index_struct_type tree --v2_path tree_v2.json
from llama_index import ComposableGraph

graph = ComposableGraph.build_from_indices(
    [index_a, index_b],
    summaries=["Summary of index A", "Summary of index B"]
)
from llama_index import download_loader

SimpleWebPageReader = download_loader('SimpleWebPageReader', use_gpt_index_import=True)
from llama_index import SimpleDirectoryReader

reader = SimpleDirectoryReader(file_paths=['docs/overview.pdf', 'docs/changelog.md'])
documents = reader.load_data()

Summary

LlamaIndex is an open-source data framework that indexes and queries information from custom data sources, intended for application developers integrating LLMs with private knowledge. As a library, it is imported into existing code, and its documentation positions it alongside frameworks for building AI agents. The project shows steady activity, with recent development contributions.

LlamaIndex is the leading document agent and OCR platform

What LlamaIndex answers

What kinds of inputs does it index from?

developers can point it at specific data sources to build an index

What architectural style does it fit into?

it operates as a library that developers import into their existing codebases

What is its output?

it produces structured data ready for querying via an index

What does it integrate with?

it works with large language models and custom data sources

Where does its control end?

it functions as a framework layer within an application rather than a standalone service

Command line

composed command

option surface across analysed releases

options at that release added, per release removed, per release

option	found in	since	description

Values are placeholders taken from each option’s declared default. Nothing is executed here — the output shown is a recording of a run that already happened.

Release history

v0.14.24 Aug 19, 2026 · issue 002

LlamaIndex v0.14.24 adds Claude Opus 5/Sonnet 5, AG-UI multimodal input, async LLMRerank, and expanded VertexAI V2 API support.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.14.24 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.14.24
```
- ›Adds raise_on_error parameter to LLM path extractors in property_graph to surface extraction failures instead of silently swallowing them.
- ›Allows Memory to accept any AsyncDBChatStore, enabling async-native database-backed chat memory stores.
- ›Migrates llama-index-tools-mcp to MCP 2.x protocol.
- ›Expands VertexAIVectorStore with V2 API support via llama-index-vector-stores-vertexaivectorsearch.
- ›Adds Claude Sonnet 5 to the llama-index-llms-anthropic and llama-index-llms-bedrock-converse model allowlists.
+6 moreshow less
- ›Adds Claude Opus 5 to the llama-index-llms-anthropic and llama-index-llms-bedrock-converse model allowlists.
- ›Supports thinking type 'disabled' in llama-index-llms-bedrock-converse.
- ›Adds GPT-5.6 models to supported models in llama-index-llms-openai.
- ›Sets Gemini 3.7 Flash as the default model in llama-index-llms-google-genai.
- ›Implements async support for LLMRerank, enabling non-blocking reranking pipelines.
- ›Supports multimodal user input (images, audio, video, documents) in the AG-UI protocol integration (llama-index-protocols-ag-ui).
v0.14.23 Jun 24, 2026 · issue -056

LlamaIndex v0.14.23 adds multimodal query engines, multimodal synthesis, and a tool-calling mock LLM to llama-index-core.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.14.23 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.14.23
```
- ›Adds multimodal query engines to llama-index-core, enabling queries that span text, image, video, and document modalities.
- ›Adds multimodal synthesis (part 2) to llama-index-core for richer cross-modal response generation.
- ›Adds a tool-calling mock LLM to llama-index-core for testing agent and tool-use pipelines without a live model.
- ›Preserves URL-backed video and document memory blocks in llama-index-core so multimodal context survives across conversation turns.
- ›Uses a set instead of a list for within-batch deduplication in the ingestion pipeline, unlocking higher-throughput document ingestion at scale.
v0.14.19 Mar 25, 2026 · issue -147

LlamaIndex v0.14.19 adds MiniMax LLM, Azure OpenAI Responses API support, LiteLLM custom providers, and GPT-5.4 Mini/Nano variants.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.14.19 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.14.19
```
└──▷ USE IT
Use the new MiniMax LLM provider for inference in a LlamaIndex pipeline.
python
```
from llama_index.llms.minimax import MiniMax

llm = MiniMax()  # defaults to M2.7
response = llm.complete('Summarize the OWASP Top 10 for 2025.')
print(response)
```
Route LiteLLM calls through a custom LLM provider by passing it in model kwargs.
python
```
from llama_index.llms.litellm import LiteLLM

llm = LiteLLM(
    model='openai/gpt-4o',
    model_kwargs={'custom_llm_provider': 'azure'}
)
response = llm.complete('List the top cloud misconfigurations.')
print(response)
```
- ›Adds llama-index-llms-minimax integration (v0.1.0) with MiniMax LLM provider, defaulting to the M2.7 model.
- ›Adds Azure OpenAI Responses API support in llama-index-llms-azure-openai.
- ›Adds support for a custom LLM provider via model kwargs in llama-index-llms-litellm.
- ›Adds support for Mini and Nano variants of GPT-5.4 in llama-index-llms-openai.
- ›Updates llama-index-llms-google-genai to default to Gemini 3 and exposes temperature control.
+1 moreshow less
- ›Enables llama-cloud>1.0 install compatibility in llama-index-core and llama-index-indices-managed-llama-cloud.
v0.14.18 Mar 16, 2026 · issue -156

LlamaIndex v0.14.18 aligns text-match filters across vector backends and expands Bedrock Claude context windows to 1M tokens.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.14.18 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.14.18
```
- ›Aligns text match filters across llama-index-core and vector store backends for consistent filter behavior (#20883).
- ›Sets context window size to 1M tokens for Claude Opus 4.6 and Sonnet 4.6 in llama-index-llms-bedrock-converse.
v0.14.16 Mar 10, 2026 · issue -161

LlamaIndex v0.14.16 adds token-bucket and sliding-window rate limiters, a multimodal reranker, GPT-5 and reasoning_content support, a ModelsLab LLM integration, and richer OpenTelemetry tracing.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.14.16 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.14.16
```
└──▷ USE IT
Use a custom embedding model inside the semantic double-merging splitter instead of the global default.
python
```
from llama_index.core.node_parser import SemanticDoubleMergingSplitterNodeParser
from llama_index.embeddings.openai import OpenAIEmbedding

parser = SemanticDoubleMergingSplitterNodeParser(
    embed_model=OpenAIEmbedding(model="text-embedding-3-small")
)
```
Introspect the schema of a large Neo4j database by sampling with APOC rather than scanning all nodes.
python
```
from llama_index.graph_stores.neo4j import Neo4jGraphStore

graph_store = Neo4jGraphStore(
    username="neo4j",
    password="<password>",
    url="bolt://localhost:7687",
    apoc_sample=0.1,
)
```
- ›Adds SlidingWindowRateLimiter to llama-index-core for strict per-minute API call caps on LLM and embedding requests.
- ›Adds token-bucket rate limiter to llama-index-core for LLM and embedding API calls.
- ›Adds optional embed_model parameter to SemanticDoubleMergingSplitterNodeParser so callers can supply a custom embedding model for semantic chunking.
- ›Adds apoc_sample parameter to llama-index-graph-stores-neo4j for sampling-based schema introspection on large Neo4j databases.
- ›Adds extra span processors via llama-index-observability-otel, enabling registration of additional processors within the OTel tracer.
+7 moreshow less
- ›Supports passing a custom tracer provider in llama-index-observability-otel.
- ›Adds inheritance for external OTel context in llama-index-observability-otel.
- ›New MultimodalLLMReranker in llama-index-core enables reranking with multimodal LLMs.
- ›Extends vector store metadata filters in llama-index-core.
- ›New llama-index-llms-modelslab integration adds ModelsLab as an LLM provider.
- ›Adds GPT-5 chat model support (gpt-5) in llama-index-llms-openai.
- ›Supports reasoning_content field in OpenAI Chat Completions responses via llama-index-llms-openai.
v0.14.15 Feb 18, 2026 · issue -180

LlamaIndex v0.14.15 adds multimodal prompt templates, AgentMesh trust layer, LayoutIR reader, OCI streaming, and more integrations.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.14.15 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.14.15
```
- ›Adds llama-index-agent-agentmesh [0.1.0], a new trust layer integration for LlamaIndex agents via AgentMesh.
- ›Adds multimodal prompt templates and a multimodal chat prompt helper to llama-index-core, enabling template variable formatting across multimodal types.
- ›Adds retry and error handling to BaseExtractor in llama-index-core.
- ›Adds support for the /predictWithStream endpoint in llama-index-llms-oci-data-science [1.0.0] for streaming use cases.
- ›Adds support for custom span processors in llama-index-observability-otel [0.3.0], with improved OpenTelemetry data serialization via dict flattening; refactored to use llama-index-instrumentation instead of llama-index-core.
+7 moreshow less
- ›Sandboxes LLM-generated code execution in EvaporateExtractor within llama-index-program-evaporate.
- ›Enhances GitHubRepoReader in llama-index-readers-github [0.10.0] with selective file fetching and deduplication.
- ›Adds pagination support for Microsoft Graph API calls in llama-index-readers-microsoft-sharepoint [0.8.0].
- ›Adds partial_params propagation to get_tools_from_mcp utils in llama-index-tools-mcp [0.4.7].
- ›Adds Claude Sonnet 4.6 model support to llama-index-llms-anthropic [0.10.9] and llama-index-llms-bedrock-converse [0.12.10].
- ›Adds Azure SDK support to llama-index-llms-mistralai [0.10.0].
- ›Adds recursive LLM type support to llama-index-core.
└──▷ BREAKING ON UPGRADE
- !The persistent_connection parameter is removed from llama-index-embeddings-ibm [0.6.0.post1] and llama-index-llms-ibm [0.7.0.post1]; any configuration referencing this parameter will break on upgrade.
- !The metadata_seperator field is removed from TextNode in llama-index-core; code or serialized objects referencing this field will break on upgrade.
v0.14.14 Feb 10, 2026 · issue -188

LlamaIndex v0.14.14 adds TokenBudgetHandler, MCP discovery, Chonkie node parser, adaptive thinking in Bedrock, and more new integrations.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.14.14 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.14.14
```
└──▷ USE IT
Pre-bind fixed parameters to an MCP tool so callers never need to supply them explicitly.
python
```
from llama_index.tools.mcp import McpToolSpec

tool_spec = McpToolSpec(
    server_url='http://localhost:8080',
    partial_params={'environment': 'production', 'tenant_id': 'acme'}
)
tools = tool_spec.to_tool_list()
```
- ›Adds TokenBudgetHandler to llama-index-core callbacks for LLM cost governance.
- ›Adds partial_params support to McpToolSpec in llama-index-tools-mcp, allowing default parameter values to be pre-bound to MCP tool calls.
- ›New llama-index-tools-mcp-discovery integration package for MCP server discovery.
- ›New llama-index-node-parser-chonkie integration package adding Chonkie as a node parser.
- ›Adds custom base_url support to the Cohere LLM integration (llama-index-llms-cohere).
+11 moreshow less
- ›Adds support for adaptive thinking in llama-index-llms-bedrock-converse.
- ›Adds support for Claude Opus 4.6 in llama-index-llms-bedrock-converse and llama-index-llms-anthropic.
- ›Adds support for gpt-5.2-chat model in llama-index-llms-openai.
- ›Adds new reasoning types in llama-index-llms-openai.
- ›Adds OpenAI-like server mode for VllmServer in llama-index-llms-vllm.
- ›Adds event and memory record deletion methods to llama-index-memory-bedrock-agentcore.
- ›Adds Sharepoint page support events to llama-index-readers-microsoft-sharepoint.
- ›Adds new solar-pro3 model support to llama-index-llms-upstage.
- ›New llama-index-tools-moss integration package adding Moss search engine as a tool.
- ›Adds LangChain 1.x support across llama-index-core, llama-index-llms-langchain, and llama-index-readers-obsidian.
- ›Makes transformers an optional dependency in llama-index-llms-openai-like and llama-index-llms-openrouter.
v0.14.13 Jan 21, 2026 · issue -208

LlamaIndex v0.14.13 adds Ray distributed ingestion, multimodal memory, new LLM/reader/vector-store integrations, and expanded agent controls.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.14.13 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.14.13
```
└──▷ USE IT
Distribute large-scale document ingestion across a Ray cluster to process corpora that are too slow to ingest on a single machine.
python
```
from llama_index.ingestion.ray import RayIngestionPipeline

pipeline = RayIngestionPipeline(
    transformations=[...],
    vector_store=my_vector_store,
)
pipeline.run(documents=my_documents)
```
Control how an agent workflow stops early when a termination condition is met, avoiding unnecessary LLM calls.
python
```
from llama_index.core.agent.workflow import AgentWorkflow

workflow = AgentWorkflow(
    agents=[...],
    early_stopping_method="generate",
)
```
- ›Adds early_stopping_method parameter to agent workflows in llama-index-core.
- ›Adds token-based code splitting support to CodeSplitter in llama-index-core.
- ›Adds configurable empty response message to synthesizers in llama-index-core.
- ›Adds milvus_partition_name parameter to add/delete operations in llama-index-vector-stores-milvus.
- ›New RayIngestionPipeline integration (llama-index-ingestion-ray v0.1.0) for distributed data ingestion.
+16 moreshow less
- ›New llama-index-readers-datasets v0.1.0 integration adds a HuggingFace Datasets reader.
- ›New llama-index-tools-parallel-web-systems v0.1.0 adds Parallel Web System tools.
- ›New llama-index-vector-stores-alibabacloud-mysql v0.1.0 adds Alibaba Cloud MySQL vector store integration.
- ›New llama-index-vector-stores-volcenginemysql v0.2.0 adds Volcengine MySQL vector store integration.
- ›New llama-index-llms-apertis v0.1.0 adds Apertis LLM integration.
- ›New multi-modal version of the Condensed Conversation & Context memory added to llama-index-core.
- ›Replaces ChatMemoryBuffer with Memory in llama-index-core.
- ›Adds support for ARNs when specifying Bedrock embedding models in llama-index-embeddings-bedrock.
- ›Adds voyage-4 models to llama-index-embeddings-voyageai.
- ›Enhances structured predict methods for Anthropic in llama-index-llms-anthropic.
- ›Adds provider routing support to llama-index-llms-openrouter.
- ›Adds hybrid search support to llama-index-vector-stores-vertexaivectorsearch.
- ›Adds Qdrant search params support to llama-index-vector-stores-qdrant.
- ›Revamps YouRetriever integration in llama-index-retrievers-you v1.0.0.
- ›Updates PatentsView reader API in llama-index-readers-patentsview v1.0.0.
- ›Improves Ollama batch embedding in llama-index-embeddings-ollama.
└──▷ BREAKING ON UPGRADE
- !ChatMemoryBuffer is replaced by Memory in llama-index-core — code instantiating ChatMemoryBuffer will need to migrate to Memory.
- !The Milvus partition parameter is renamed to milvus_partition_name in add/delete — callers using the old parameter name will break.
- !llama-index-llms-gemini is deprecated in v0.6.2.
v0.14.12 Dec 30, 2025 · issue -230

LlamaIndex v0.14.12 adds async tool spec support, Element node parser, new LLM/embedding integrations, and MongoDB async Atlas support.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.14.12 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.14.12
```
└──▷ USE IT
Keep an Ollama embedding model loaded in memory between requests to avoid cold-start latency in high-throughput pipelines.
python
```
from llama_index.embeddings.ollama import OllamaEmbedding

embed_model = OllamaEmbedding(
    model_name="nomic-embed-text",
    keep_alive="10m",
)
embeddings = embed_model.get_text_embedding("Hello, world!")
```
- ›Adds keep_alive parameter to llama-index-embeddings-ollama embedding class to control model persistence in memory.
- ›Switches use_file_api to a flexible file_mode field in llama-index-llms-google-genai for more granular file upload handling, with a bump to google-genai v1.52.0.
- ›Adds gpt-5.2 and gpt-5.2 pro model support to llama-index-llms-openai.
- ›Adds async support to ToolSpec across llama-index-core, llama-index-vector-stores-azurepostgresql, llama-index-vector-stores-lancedb, and llama-index-callbacks-agentops.
- ›Adds new Element node parser to llama-index-core for structured element-level document parsing.
+11 moreshow less
- ›Adds new llama-index-llms-aibadgr integration (v0.1.0) for AI Badgr OpenAI-compatible LLMs.
- ›Adds new llama-index-tools-typecast integration (v0.1.0) with text-to-speech features.
- ›Adds MENTIONS edge type to the NebulaGraph property graph store in llama-index-graph-stores-nebula.
- ›Adds Voyage Multimodal 3.5 model support to llama-index-embeddings-voyageai.
- ›Adds async MongoDB Atlas vector store support to llama-index-vector-stores-mongodb.
- ›Adds delete index capability to llama-index-vector-stores-mongodb.
- ›Adds Google Vertex AI Vector Search v2.0 support to llama-index-vector-stores-vertexaivectorsearch.
- ›Permits passing a custom httpx.AsyncClient when constructing a BasicMCPClient in llama-index-tools-mcp.
- ›Restores haiku-3 model support to llama-index-llms-anthropic.
- ›Improves MockFunctionCallingLLM in llama-index-core for better testing of function-calling workflows.
- ›Adds positional thought signature for 'thoughts' in llama-index-llms-google-genai.
└──▷ BREAKING ON UPGRADE
- !The use_file_api field in llama-index-llms-google-genai is replaced by file_mode; existing code setting use_file_api must be updated to use file_mode.
v0.14.10 Dec 4, 2025 · issue -256

LlamaIndex v0.14.10 adds a mock function-calling LLM for testing and a new Airweave tool integration with advanced search.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.14.10 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.14.10
```
- ›Adds llama-index-tools-airweave (v0.1.0) integration, enabling agents to use Airweave's advanced search features as a tool.
- ›Adds a mock function-calling LLM to llama-index-core for testing agent and tool-calling pipelines without a live model.
v0.14.9 Dec 2, 2025 · issue -258

LlamaIndex v0.14.9 adds multi-modal chat engine, OVHcloud LLM provider, Bedrock inference profiles, and Claude Opus 4.5 / GPT-5.1 model support.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.14.9 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.14.9
```
- ›Adds llama-index-llms-ovhcloud integration (v0.1.0), a new LLM provider for OVHcloud AI Endpoints.
- ›Adds support for Amazon Bedrock Application Inference Profiles in llama-index-embeddings-bedrock.
- ›MultiModalVectorStoreIndex now returns a multi-modal ContextChatEngine, enabling richer multi-modal chat workflows.
- ›Adds anthropic claude opus 4.5 model support across llama-index-llms-anthropic and llama-index-llms-bedrock-converse.
- ›Adds gpt-5.1-chat model support in llama-index-llms-openai.
+1 moreshow less
- ›llama-index-readers-confluence now uses HtmlTextParser for HTML-to-Markdown conversion and is relicensed to MIT.
v0.14.8 Nov 10, 2025 · issue -280

LlamaIndex 0.14.8 adds buffer support for media blocks, ScrapyWebReader, Bedrock tool call block integration, and OpenAI v2 SDK support across packages.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.14.8 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.14.8
```
└──▷ USE IT
Load web pages at scale using the new Scrapy-based reader when you need to crawl structured or JavaScript-heavy sites.
python
```
from llama_index.readers.web import ScrapyWebReader

reader = ScrapyWebReader()
documents = reader.load_data(urls=["https://example.com"])
```
- ›Adds buffer field to image, audio, video, and document blocks in llama-index-core, enabling direct binary data handling in multimodal pipelines.
- ›Adds ScrapyWebReader integration in llama-index-readers-web, enabling Scrapy-based web crawling as a document source.
- ›Adds RawMessageDeltaEvent support in llama-index-llms-anthropic streaming responses.
- ›Integrates tool call block support into llama-index-llms-bedrock-converse, aligning Bedrock Converse with the tool-block pattern.
- ›Integrates tool block support into llama-index-llms-google-genai, aligning Google GenAI with the tool-block pattern.
+3 moreshow less
- ›Adds token usage information to additional_kwargs in llama-index-llms-google-genai chat responses.
- ›Adds OpenAI v2 SDK support across llama-index-llms-openai, llama-index-llms-upstage, llama-index-packs-streamlit-chatbot, llama-index-packs-voyage-query-engine, llama-index-readers-whisper.
- ›Updates llama-index-llms-bedrock-converse model name extraction to include the jp region prefix.
v0.14.7 Oct 30, 2025 · issue -291

LlamaIndex v0.14.7 adds SerpEx tool, GitHub App auth, Bedrock Guardrails streaming, and tool-call-block support for Anthropic, MistralAI, and Ollama.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.14.7 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.14.7
```
- ›Adds llama-index-tools-serpex (v0.1.0), a new SerpEx search tool integration for agent pipelines.
- ›Adds streamProcessingMode support for Bedrock Guardrails in llama-index-llms-bedrock-converse, enabling streaming-compatible guardrail enforcement.
- ›Adds optional streamProcessingMode for Bedrock structured output (previously forced), giving callers control over when it is applied.
- ›Adds GitHub App authentication support to llama-index-readers-github (v0.9.0), complementing existing token-based auth.
- ›Integrates tool-call-block support into llama-index-llms-anthropic (v0.10.0), llama-index-llms-mistralai (v0.9.0), and llama-index-llms-ollama (v0.9.0) for structured tool-use responses.
+5 moreshow less
- ›Updates llama-index-embeddings-voyageai (v0.5.0) with the latest VoyageAI integration.
- ›Adds Hyperscale and Composite Vector Index support to llama-index-vector-stores-couchbase (v0.6.0).
- ›Makes SVG processing optional in llama-index-readers-confluence (v0.5.0), removing the hard pycairo install requirement.
- ›Updates available models in llama-index-llms-fireworks (v0.4.5).
- ›Allows setting the temperature parameter for gpt-5-chat in llama-index-llms-openai (v0.6.6).
v0.14.6 Oct 26, 2025 · issue -295

LlamaIndex v0.14.6 adds parallel tool calls, Isaacus and Helicone integrations, async Bedrock retriever, and GIN index support for PostgreSQL vector store.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.14.6 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.14.6
```
└──▷ USE IT
Use the async Bedrock Knowledge Bases retriever to query Amazon Knowledge Bases in an async pipeline without blocking.
python
```
from llama_index.retrievers.bedrock import AmazonKnowledgeBasesRetriever

retriever = AmazonKnowledgeBasesRetriever(
    knowledge_base_id='<knowledge_base_id>',
    retrieval_config={'vectorSearchConfiguration': {'numberOfResults': 5}},
)
results = await retriever.aretrieve('What is our incident response policy?')
```
- ›Adds allow_parallel_tool_calls parameter to non-streaming tool call support in llama-index-core.
- ›Adds GIN index support for text array metadata in the PostgreSQL vector store (llama-index-vector-stores-postgres).
- ›Adds async support for AmazonKnowledgeBasesRetriever in llama-index-retrievers-bedrock.
- ›New llama-index-embeddings-isaacus integration (v0.1.0) adds Isaacus embeddings support.
- ›New llama-index-llms-helicone integration (v0.1.0) adds Helicone LLM support.
+2 moreshow less
- ›Adds GLM model support to llama-index-llms-baseten.
- ›Updates OCI GenAI Cohere models in both llama-index-embeddings-oci-genai and llama-index-llms-oci-genai.
v0.14.5 Oct 15, 2025 · issue -306

v0.14.5 adds SGLang LLM integration, SignNow MCP tools, Tavily URL extraction, Azure PostgreSQL hybrid search, and new model support across Anthropic, Bedrock, OpenAI, and Fireworks.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.14.5 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.14.5
```
└──▷ USE IT
Run inference through a local SGLang server using the new first-class SGLang LLM integration.
python
```
from llama_index.llms.sglang import SGLang

llm = SGLang(model="meta-llama/Llama-3.1-8B-Instruct")
response = llm.complete("Explain prompt injection attacks in one paragraph.")
print(response)
```
- ›Adds llama-index-llms-sglang (v0.1.0) — a new SGLang LLM integration for running local inference via SGLang.
- ›Adds llama-index-tools-signnow (v0.1.0) — a new SignNow MCP tools integration for document signing workflows.
- ›Adds a Tavily extract function in llama-index-tools-tavily-research for URL content extraction.
- ›Adds hybrid search support to llama-index-vector-stores-azurepostgresql.
- ›Adds prompt caching model validation utilities to llama-index-llms-anthropic.
+8 moreshow less
- ›Adds support for custom models in llama-index-llms-fireworks.
- ›Adds support for xAI models in llama-index-llms-oci-genai.
- ›Adds haiku 4.5 model support to llama-index-llms-anthropic and llama-index-llms-bedrock-converse.
- ›Adds Claude Sonnet 4.5 as a reasoning model and Opus 4.1 function-calling model support in llama-index-llms-bedrock-converse.
- ›Adds support for global cross-region inference profile prefix in llama-index-llms-bedrock-converse.
- ›Adds GPT-5 and GPT-5 Pro model support (including JSON_SCHEMA_MODELS) in llama-index-llms-openai.
- ›Adds pagination parameters for repository tree and issues in llama-index-readers-gitlab.
- ›Adds a progress bar for multiprocess document loading in llama-index-core.
v0.14.4 Oct 3, 2025 · issue -317

LlamaIndex v0.14.4 adds Bedrock AgentCore Memory, Apache Solr vector store, structured outputs for OpenAILike, and Claude Sonnet 4.5 support.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.14.4 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.14.4
```
- ›Adds llama-index-memory-bedrock-agentcore (v0.1.0) with a new Bedrock AgentCore Memory integration for persistent agent memory backed by AWS.
- ›Adds llama-index-vector-stores-solr (v0.1.0) with a new ApacheSolrVectorStore integration for using Apache Solr as a vector store backend.
- ›Adds structured outputs support to OpenAILike in llama-index-llms-openai-like and llama-index-llms-openai.
- ›Adds support for Claude Sonnet 4.5 (anthropic-sonnet-4-5) in llama-index-llms-anthropic.
- ›Adds support for Claude Sonnet 4.5 in llama-index-llms-bedrock-converse.
+2 moreshow less
- ›Expands the list of available models in llama-index-llms-mistralai with updated MistralAI LLM entries.
- ›Updates llama-index-tools-scrapegraph to align with the latest scrapegraphai library.
v0.14.3 Sep 24, 2025 · issue -326

LlamaIndex v0.14.3 adds ThinkingBlock content support across LLMs, a PaddleOCR reader, Azure PostgreSQL vector store, and Valyu Extractor with Fast mode.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.14.3 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.14.3
```
- ›New llama-index-readers-paddle-ocr package (PaddleOCR Reader) extracts text from images embedded in PDFs.
- ›New llama-index-vector-stores-azurepostgresql package adds vector store support for Azure PostgreSQL.
- ›Adds ThinkingBlock as a supported content block type across llama-index-core, llama-index-llms-anthropic, llama-index-llms-google-genai, llama-index-llms-mistralai, and llama-index-llms-openai.
- ›Adds Valyu Extractor and Fast mode to llama-index-tools-valyu.
- ›llama-index-llms-google-genai gains FileAPI support for document uploads, previously missing.
+3 moreshow less
- ›llama-index-readers-mongodb, llama-index-storage-chat-store-mongo, and llama-index-storage-kvstore-mongodb migrate from Motor to the PyMongo native asynchronous API.
- ›llama-index-readers-web Firecrawl integration migrates to the Firecrawl v2 SDK.
- ›llama-index-llms-baseten adds support for the kimik2-0905 model and introduces Dynamic Model APIs validation.
v0.14.0 Sep 8, 2025 · issue -342

LlamaIndex v0.14.0 adds document block support in OpenAI chat completions and upgrades to llama-index-workflows 2.0.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.14.0 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.14.0
```
- ›Adds support for document blocks in llama-index-llms-openai OpenAI chat completions.
└──▷ BREAKING ON UPGRADE
- !The llama-index-workflows dependency is bumped to 2.0: the checkpointer feature is removed, sub-workflows are removed, the send_event method is removed from the Workflow class (it remains on the Context class), the stream_events() method is removed from the Workflow class (it remains on the Context class), and stepwise execution support is removed.
v0.13.5 Sep 4, 2025 · issue -346

LlamaIndex v0.13.5 adds thinking delta in AgentStream events, system prompt/tool caching for BedrockConverse, and a YugabyteDB chat store.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.13.5 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.13.5
```
└──▷ USE IT
Stream an agent response and surface the model's thinking delta alongside the output token stream.
python
```
async for event in agent.astream_chat("Explain RSA encryption"):
    if hasattr(event, 'thinking_delta') and event.thinking_delta:
        print("[thinking]", event.thinking_delta)
    if hasattr(event, 'delta') and event.delta:
        print(event.delta, end="", flush=True)
```
Enable tool caching and a system prompt on a BedrockConverse LLM to reduce latency and token costs on repeated tool calls.
python
```
from llama_index.llms.bedrock_converse import BedrockConverse

llm = BedrockConverse(
    model="anthropic.claude-3-5-sonnet-20241022-v2:0",
    system_prompt="You are a security analyst assistant.",
    tool_caching=True,
)
```
- ›Adds thinking_delta field to AgentStream events in llama-index-core to expose thinking deltas from LLM responses.
- ›Adds system prompt and tool caching config kwargs to BedrockConverse in llama-index-llms-bedrock-converse.
- ›New llama-index-storage-chat-store-yugabytedb package (v0.1.0) introduces a YugabyteDB-backed chat store.
v0.13.4 Sep 2, 2025 · issue -348

LlamaIndex v0.13.4 adds Baseten LLM/embedding integrations, PostgreSQL schema support, MMR search for Chroma, and Qdrant payload indexes.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.13.4 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.13.4
```
└──▷ USE IT
Run MMR search against a Chroma vector store to retrieve diverse, non-redundant results for threat-intel queries.
python
```
from llama_index.vector_stores.chroma import ChromaVectorStore
from llama_index.core.vector_stores.types import VectorStoreQuery, VectorStoreQueryMode

vector_store = ChromaVectorStore(chroma_collection=collection)
query = VectorStoreQuery(
    query_embedding=embedding,
    similarity_top_k=10,
    mode=VectorStoreQueryMode.MMR,
)
results = vector_store.query(query)
```
- ›Adds schema support for PostgreSQL to Memory and SQLAlchemyChatStore, enabling multi-tenant or schema-isolated chat storage.
- ›Adds amazon.nova-premier-v1:0 to BEDROCK_MODELS in llama-index-llms-bedrock-converse.
- ›Adds MMR (Maximal Marginal Relevance) search to llama-index-vector-stores-chroma.
- ›Adds payload indexes support to QdrantVectorStore in llama-index-vector-stores-qdrant.
- ›Adds an option for an initial tool choice in FunctionAgent.
+5 moreshow less
- ›Adds a sync wrapper for put_messages in Memory.
- ›New llama-index-embeddings-baseten [0.1.0] and llama-index-llms-baseten [0.1.0] packages add Baseten as an LLM and embedding provider.
- ›Adds ZenRows web reader to llama-index-readers-web.
- ›IBM integrations (llama-index-embeddings-ibm, llama-index-llms-ibm, llama-index-postprocessor-ibm) now support additional/external URLs beyond the default endpoint.
- ›Google Drive reader now surfaces Google API errors explicitly instead of silently failing.
v0.13.3 Aug 22, 2025 · issue -359

LlamaIndex v0.13.3 adds Heroku embeddings, Qdrant sharding, instruction-enhanced Ollama embeddings, and GPT-5 model support.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.13.3 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.13.3
```
└──▷ USE IT
Point an existing OpenAI LLM config at GPT-5 to start testing the new model.
python
```
from llama_index.llms.openai import OpenAI

llm = OpenAI(model="gpt-5-chat-latest")
response = llm.complete("Summarize the MITRE ATT&CK framework in three sentences.")
print(response)
```
- ›Adds HerokuEmbeddings class in new llama-index-embeddings-heroku 0.1.0 package for embedding via Heroku-hosted models.
- ›Adds instruction support to OllamaEmbedding in llama-index-embeddings-ollama, enabling instruction-prefixed embedding requests.
- ›Adds gpt-5-chat-latest model support to llama-index-llms-openai.
- ›Adds Qdrant sharding support to llama-index-vector-stores-qdrant.
v0.13.2 Aug 14, 2025 · issue -363

LlamaIndex v0.13.2 adds streaming control in agents, Superlinked retriever, OpenAI-OSS models on Bedrock, enhanced PowerPoint extraction, and MCP custom type handlers.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.13.2 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.13.2
```
- ›Adds support for disabling streaming in agents (llama-index-core 0.13.2).
- ›Adds llama-index-retrievers-superlinked 0.1.0, a new Superlinked retriever integration.
- ›Adds OpenAI-OSS models to BedrockConverse in llama-index-llms-bedrock-converse 0.8.2.
- ›Enhances the PowerPoint reader (llama-index-readers-file 0.5.1) with comprehensive content extraction.
- ›Adds handlers for custom types and Pydantic models in MCP tools (llama-index-tools-mcp 0.4.0).
+1 moreshow less
- ›Updates llama-index-vector-stores-clickhouse 0.6.0 with new vector search capabilities from ClickHouse.
v0.13.1 Aug 8, 2025 · issue -363

LlamaIndex v0.13.1 adds Heroku LLM integration, Bedrock AgentCore toolspecs, voyage context embeddings, BM25 metadata filtering, and more.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.13.1 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.13.1
```
- ›Adds metadata filtering support to BM25Retriever (via llama-index-retrievers-bm25 0.6.2), enabling filtered keyword retrieval alongside vector stores.
- ›Adds llama-index-llms-heroku 0.1.0, a new LLM integration for Heroku-hosted models.
- ›Adds llama-index-tools-aws-bedrock-agentcore 0.1.0 with toolspecs for Bedrock AgentCore browser and code interpreter.
- ›Adds voyage context embeddings support to llama-index-embeddings-voyageai 0.4.1.
- ›Adds Anthropic citations to non-beta (GA) support in llama-index-llms-anthropic 0.8.2.
+6 moreshow less
- ›Adds support for gpt-5 in llama-index-llms-openai 0.5.2.
- ›Adds support for gpt-oss NIM in llama-index-llms-nvidia 0.4.1.
- ›Enables partially formatted system prompts for the ReAct agent in llama-index-core 0.13.1.
- ›Adds support for presidio entities in llama-index-postprocessor-presidio 0.5.0.
- ›Updates Kuzu graph store integration to the latest SDK in llama-index-graph-stores-kuzu 0.9.0.
- ›Allows top_k values greater than the number of indexed nodes in BM25Retriever.
v0.13.0 Jul 31, 2025 · issue -364

LlamaIndex v0.13.0 overhauls agents, adds Gemini Live voice, and expands vector store and reader capabilities.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.13.0 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.13.0
```
└──▷ USE IT
Access an S3 bucket in a specific AWS region when loading documents with S3Reader.
python
```
from llama_index.readers.s3 import S3Reader

reader = S3Reader(
    bucket="my-bucket",
    client_kwargs={"region_name": "eu-west-1"}
)
documents = reader.load_data()
```
Build a multi-step reasoning agent using the new workflow-based API after migrating off deprecated agent classes.
python
```
from llama_index.core.agent.workflow import FunctionAgent

agent = FunctionAgent(
    tools=[my_tool],
    llm=llm,
    system_prompt="You are a helpful assistant."
)
response = await agent.run("What is the capital of France?")
```
- ›Adds partition_names parameter to Milvus search configuration in llama-index-vector-stores-milvus for scoped partition-level queries.
- ›Adds client_kwargs support (including region_name) to S3Reader in llama-index-readers-s3 for region-aware S3 access.
- ›Adds get-nodes and delete-nodes operations to llama-index-vector-stores-astradb.
- ›Adds ANY/ALL postgres operator support to llama-index-vector-stores-postgres.
- ›Adds file filtering and custom processing enhancements to llama-index-readers-github.
+8 moreshow less
- ›Adds Thought Summaries and signatures support for Gemini in llama-index-llms-google-genai.
- ›Adds support for kimi-k2-instruct model in llama-index-llms-nvidia.
- ›Adds solar-pro2 model support to llama-index-llms-upstage.
- ›Introduces first beta implementation of Gemini Live in llama-index-voice-agents-gemini-live.
- ›Updates mixedbread embeddings (llama-index-embeddings-mixedbreadai) and reranker (llama-index-postprocessor-mixedbreadai-rerank) for the latest SDK.
- ›Updates Valyu SDK integration to latest version in llama-index-tools-valyu.
- ›Replaces legacy agent classes with new workflow-based agents: FunctionAgent, CodeActAgent, ReActAgent, and AgentWorkflow in llama-index-core.
- ›Changes default index.as_chat_engine() to return a CondensePlusContextChatEngine in llama-index-core.
└──▷ BREAKING ON UPGRADE
- !Removed deprecated agent classes FunctionCallingAgent, the older ReActAgent implementation, AgentRunner, all step workers, StructuredAgentPlanner, and OpenAIAgent from llama-index-core; migrate to FunctionAgent, CodeActAgent, ReActAgent, or AgentWorkflow.
- !Removed deprecated QueryPipeline class and all associated code from llama-index-core.
- !index.as_chat_engine() now returns a CondensePlusContextChatEngine by default; agent-based chat engines have been removed.
v0.12.52 Jul 23, 2025 · issue -364

LlamaIndex v0.12.52 adds a Jira issue tool spec, web reader timeouts, and optimized BGEM3Index persistence.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.12.52 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.12.52
```
- ›Adds timeout parameter to webpage readers in llama-index-readers-web, defaulting to 60 seconds.
- ›New llama-index-tools-jira-issue package (v0.1.0) introducing a Jira issue tool spec for agent use.
- ›Optimizes memory usage for BGEM3Index persistence in llama-index-indices-managed-bge-m3.
v0.12.51 Jul 22, 2025 · issue -364

FunctionTool gains auto type conversion for basic Python types like date when using Pydantic fields.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.12.51 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.12.51
```
- ›Enhances FunctionTool with automatic type conversion for basic Python types (e.g., date) when declared as Pydantic fields in tool functions.
v0.12.50 Jul 19, 2025 · issue -364

LlamaIndex v0.12.50 adds Cloudflare AI Gateway LLM, S3 vector store, ServiceNow reader, HTML table extraction, and Google Search tool support.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.12.50 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.12.50
```
- ›Adds google_search tool support to the llama-index-llms-google-genai GoogleGenAI LLM integration.
- ›Introduces llama-index-llms-cloudflare-ai-gateway [0.1.0], a new LLM integration for Cloudflare AI Gateway.
- ›Introduces llama-index-vector-stores-s3 [0.1.0] with S3 Vectors support as a new vector store backend.
- ›Adds llama-index-readers-service-now [0.1.0], a new reader for ServiceNow data.
- ›Adds HTML table extraction support to MarkdownElementNodeParser in llama-index-core.
+2 moreshow less
- ›Improves instrumentation span naming in llama-index-instrumentation [0.3.0].
- ›Adds Llama 4 models to llama-index-llms-bedrock-converse; removes Llama 3.2 1B and 3B from function-calling models.
└──▷ BREAKING ON UPGRADE
- !The get_cache_dir() function in llama-index-core changes its default cache directory location to a more secure path — existing setups relying on the previous default location may need to update their configuration or migrate cached data.
- !llama-index-llms-bedrock-converse: Llama 3.2 1B and 3B models are removed from the list of supported function-calling models.
v0.12.49 Jul 14, 2025 · issue -364

LlamaIndex v0.12.49 adds structured output in agents, DuckDB stores, Moorcheh vector store, and retry for workflow agents.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.12.49 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.12.49
```
- ›Adds retry capability to workflow agents in llama-index-core.
- ›Adds structured output support in agents (llama-index-core) as a first implementation.
- ›Adds llama-index-storage-kvstore-duckdb [0.1.3], llama-index-storage-docstore-duckdb [0.1.0], and llama-index-storage-index-store-duckdb [0.1.0] packages, providing DuckDB-backed KV, document, and index stores.
- ›Adds async support and faster cosine similarity to llama-index-vector-stores-duckdb.
- ›Adds llama-index-vector-stores-moorcheh [0.1.0] with a new Moorcheh vector store integration.
+2 moreshow less
- ›Adds support in llama-index-llms-nvidia to use LLM models outside the default list.
- ›Adds RetrieverQueryEngine async node postprocessor support in llama-index-core.
v0.12.48 Jul 9, 2025 · issue -364

LlamaIndex v0.12.48 adds cached content support for GoogleGenAI and image prompt support for OCI Generative AI Llama models.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.12.48 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.12.48
```
- ›Adds cached content support to the llama-index-llms-google-genai integration (v0.2.4), enabling reuse of cached context in GoogleGenAI LLM calls.
- ›Adds image prompt support for OCI Generative AI Llama models in llama-index-llms-oci-genai (v0.5.1).
- ›Reduces trips to the KV store during Document Hash Checks in llama-index-core, improving performance for large document ingestion workflows.
v0.12.47 Jul 7, 2025 · issue -364

LlamaIndex v0.12.47 adds agent iteration limits, forced tool calling, Anthropic citations, LanceDB multimodal integration, and OCI GenAI image prompts.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.12.47 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.12.47
```
└──▷ USE IT
Cap an agent's reasoning loop to prevent infinite tool calls in production workflows.
python
```
result = agent.run('Summarize the top 5 findings from this report', max_iterations=20)
```
- ›Adds default max_iterations argument (value: 20) to the .run() method on agents in llama-index-core, capping runaway agent loops out of the box.
- ›Sets tool_required=True by default in FunctionCallingProgram and structured LLMs where supported, ensuring tool calls are always attempted rather than optionally skipped.
- ›New Anthropic citations support in llama-index-llms-anthropic v0.7.6.
- ›Adds image prompt support for OCI Generative AI Llama models in llama-index-llms-oci-genai.
- ›New llama-index-indices-managed-lancedb v0.1.0 integration for LanceDB MultiModal AI LakeHouse.
+2 moreshow less
- ›Base LLM classes in llama-index-core now support multi-modal features natively via ImageBlock, replacing the former dedicated Multi Modal LLM classes.
- ›Adds Firecrawl as an integration source in llama-index-readers-web.
└──▷ BREAKING ON UPGRADE
- !Multi Modal LLMs are deprecated in llama-index-core; all existing multi-modal LLM classes are now extensions of their base LLM counterpart, which handles multi-modal features internally via ImageBlock.
v0.12.46 Jul 3, 2025 · issue -364

LlamaIndex v0.12.46 adds async delete and insert methods to VectorStoreIndex.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.12.46 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.12.46
```
- ›Adds async delete and insert methods to VectorStoreIndex in llama-index-core, enabling non-blocking vector store mutations in async workflows.
v0.12.45 Jul 1, 2025 · issue -364

LlamaIndex v0.12.45 adds tool content block output, chat UI events, AWS Bedrock Claude models, and async Google Search support.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.12.45 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.12.45
```
└──▷ USE IT
Constrain the dimensionality of Azure OpenAI embeddings to reduce storage and speed up similarity search.
python
```
from llama_index.embeddings.azure_openai import AzureOpenAIEmbedding

embed_model = AzureOpenAIEmbedding(
    model="text-embedding-3-large",
    deployment_name="my-embedding-deployment",
    dimensions=512,
    azure_endpoint="https://<your-resource>.openai.azure.com/",
    api_key="<your-api-key>",
)
```
- ›Adds dimensions parameter to AzureOpenAIEmbedding in llama-index-embeddings-azure-openai for controlling embedding output size.
- ›Allows tools to output content blocks in llama-index-core, enabling richer structured tool responses.
- ›Adds chat UI events and models to the llama-index-core package.
- ›Adds new AWS Claude models available on Bedrock to llama-index-llms-anthropic.
- ›Adds proper async Google Search support to GoogleSearchToolSpec in llama-index-tools-google.
+1 moreshow less
- ›Adapts llama-index-memory-mem0 to the new framework memory standard.
v0.12.44 Jun 26, 2025 · issue -365

LlamaIndex v0.12.44 adds IBM Db2 vector store, OpenAI Realtime Conversation, CachePoint chat blocks, and Pinecone v7 support.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.12.44 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.12.44
```
└──▷ USE IT
Cache an expensive system or user message turn to avoid recomputing token context on repeated LLM calls.
python
```
from llama_index.core.llms import ChatMessage
from llama_index.core.base.llms.types import CachePoint

messages = [
    ChatMessage(role="user", content=[
        {"type": "text", "text": "You are a helpful assistant with a large knowledge base."},
        CachePoint(),
    ])
]
```
Pass advanced cross-encoder options (e.g. a custom device or batch size) when reranking with SBERT.
python
```
from llama_index.postprocessor.sbert_rerank import SentenceTransformerRerank

reranker = SentenceTransformerRerank(
    model="cross-encoder/ms-marco-MiniLM-L-6-v2",
    top_n=5,
    cross_encoder_kwargs={"device": "cuda", "max_length": 512},
)
```
- ›Adds CachePoint content block to llama-index-core for caching chat messages in conversations.
- ›Adds cross_encoder_kwargs parameter to llama-index-postprocessor-sbert-rerank for advanced cross-encoder configuration.
- ›Enables forwarding of arbitrary Azure Search SDK parameters in AzureAISearchVectorStore for document retrieval.
- ›New llama-index-vector-stores-db2 package (v0.1.0) adds IBM Db2 as a supported vector store.
- ›Adds batch support for llama-index-embeddings-fastembed.
+5 moreshow less
- ›Adds async batching for llama-index-embeddings-huggingface using asyncio.to_thread.
- ›Refactors DuckDB VectorStore in llama-index-vector-stores-duckdb (v0.4.0).
- ›Supports Pinecone v7 in llama-index-vector-stores-pinecone (v0.6.0).
- ›Adds beta OpenAI Realtime Conversation integration via new llama-index-voice-agents-openai package.
- ›Adds visualization functions for single and multi-agent workflows in llama-index-utils-workflow.
v0.12.43 Jun 19, 2025 · issue -365

LlamaIndex v0.12.43 adds ag-ui protocol, openGauss vector store, Hive Intelligence search tool, async MongoDB reader, and mermaid workflow diagrams.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.12.43 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.12.43
```
└──▷ USE IT
Visualise a workflow's structure as a mermaid diagram for documentation or debugging.
python
```
from llama_index.utils.workflow import draw_all_possible_flows

draw_all_possible_flows(MyWorkflow, filename="workflow.html")
```
- ›Adds llama-index-protocols-ag-ui package with ag-ui protocol support for agentic UI integrations.
- ›Adds llama-index-vector-stores-opengauss [0.1.0] with openGauss vector store integration.
- ›Adds llama-index-tools-hive [0.1.0] with a Hive Intelligence search tool.
- ›Adds async driver support via alazy_load_data to llama-index-readers-mongodb.
- ›Adds cache_dir parameter to the Sentence Transformers post-processor in llama-index-postprocessor-sbert-rerank.
+5 moreshow less
- ›Moves Workflows code out to its own llama-index-workflows package (with backward compatibility retained in core).
- ›Moves instrumentation code out to its own llama-index-instrumentation package.
- ›Makes BaseWorkflowAgent a workflow itself, enabling it to be composed directly as a workflow.
- ›Adds mermaid diagram drawing support for workflows in llama-index-utils-workflow.
- ›Improves robustness of the llama-index-llms-perplexity integration.
v0.12.42 Jun 12, 2025 · issue -365

LlamaIndex v0.12.42 adds reasoning support for Mistral/Magistral, OpenAI o3-pro, a multimodal OpenAI-like LLM package, figure retrieval, and an ArtifactEditorToolSpec.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.12.42 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.12.42
```
- ›New llama-index-tools-artifact-editor [0.1.0] package introduces ArtifactEditorToolSpec for editing Pydantic objects as a tool.
- ›New llama-index-multi-modal-llms-openai-like [0.1.0] package adds an OpenAI-compatible multi-modal LLM integration.
- ›Adds reasoning support (including Magistral) to llama-index-llms-mistralai [0.6.0].
- ›Adds day-0 support for OpenAI o3-pro in llama-index-llms-openai [0.4.5].
- ›Adds figure retrieval SDK integration to llama-index-indices-managed-llama-cloud [0.7.7].
+3 moreshow less
- ›Adds the ability to exclude source fields from query responses in llama-index-vector-stores-opensearch [0.5.6].
- ›Adds label truncation to workflow visualization in llama-index-utils-workflow [0.3.3].
- ›llama-index-postprocessor-bedrock-rerank [0.3.3] prefers BedrockRerank as the canonical class name over AWSBedrockRerank.
v0.12.41 Jun 7, 2025 · issue -365

LlamaIndex v0.12.41 adds ApertureDB property graph, ElevenLabs voice agents, Ollama thinking, OpenAI JSON Schema output, and Milvus upsert support.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.12.41 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.12.41
```
- ›Adds MutableMappingKVStore to llama-index-core for easier in-process caching backed by any MutableMapping implementation.
- ›Adds DocumentBlock support to the LiteLLM LLM integration (llama-index-llms-litellm 0.5.1), enabling multimodal document inputs through LiteLLM.
- ›Adds support for Ollama's think feature in llama-index-llms-ollama 0.6.2, exposing model chain-of-thought reasoning.
- ›Adds OpenAI JSON Schema structured output support in llama-index-llms-openai 0.4.4.
- ›Adds log recording during MCP tool calls in llama-index-tools-mcp 0.2.5.
+5 moreshow less
- ›Adds upsert entities support to llama-index-vector-stores-milvus 0.8.4.
- ›New llama-index-graph-stores-ApertureDB 0.1.0 package introduces ApertureDB as a property graph store.
- ›New llama-index-voice-agents-elevenlabs 0.1.0-beta package adds ElevenLabs voice agent integration.
- ›New llama-index-packs-searchain 0.1.0 package adds the Searchain LlamaPack.
- ›Allows newer versions of gcsfs in llama-index-readers-gcs 0.4.1, unblocking dependency upgrades.
└──▷ BREAKING ON UPGRADE
- !JsonPickleSerializer is renamed to PickleSerializer in llama-index-core.
v0.12.40 Jun 3, 2025 · issue -365

LlamaIndex v0.12.40 adds StopEvent validation, static AWS credentials for Anthropic Bedrock, a Measure Space tool pack, and MCP client header support.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.12.40 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.12.40
```
└──▷ USE IT
Authenticate MCP client requests by passing custom headers — useful when your MCP server requires an API key or auth token.
python
```
from llama_index.tools.mcp import BasicMCPClient

client = BasicMCPClient(
    url="https://my-mcp-server.example.com",
    headers={"Authorization": "Bearer <token>"}
)
```
- ›Adds header handling to BasicMCPClient in llama-index-tools-mcp, enabling authenticated MCP connections.
- ›New llama-index-tools-measurespace [0.1.0] package adds weather, climate, air quality, and geocoding tools from Measure Space.
- ›Supports passing static AWS credentials to Anthropic Bedrock via llama-index-llms-anthropic.
- ›Enforces StopEvent step validation in llama-index-core workflows so only one step can handle a StopEvent.
v0.12.39 May 30, 2025 · issue -366

LlamaIndex v0.12.39 adds Workflow dependency injection, tool_required for function-calling LLMs, and multi-language Milvus analyzer support.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.12.39 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.12.39
```
- ›Adds tool_required param to function-calling LLMs in llama-index-core, letting callers force the model to invoke a tool rather than return plain text.
- ›Introduces a Resource primitive to llama-index-core Workflows for structured dependency injection across workflow steps.
- ›Adds multi-language analyzer support in llama-index-vector-stores-milvus (v0.8.3), enabling language-aware tokenization for Milvus full-text search.
- ›Adds non-persisted composite retrieval to llama-index-indices-managed-llama-cloud (v0.7.2) for in-memory combined index queries without writing to LlamaCloud.
- ›Updates llama-index-llms-ollama (v0.6.1) to support the Ollama 0.5.0 SDK.
v0.12.38 May 29, 2025 · issue -366

LlamaIndex v0.12.38 adds embeddings caching, Claude 4, OpenTelemetry observability, Azure Foundry agent, and overhauled MCP client support.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.12.38 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.12.38
```
└──▷ USE IT
Enable parallel tool calls in a FunctionAgent to allow the LLM to invoke multiple tools concurrently in a single step.
python
```
from llama_index.core.agent import FunctionAgent

agent = FunctionAgent(
    tools=[...],
    llm=llm,
    allow_parallel_tool_calls=True,
)
```
Configure NLSQLTableQueryEngine with separate row, column, and table retrievers for fine-grained SQL retrieval control.
python
```
from llama_index.core.query_engine import NLSQLTableQueryEngine

query_engine = NLSQLTableQueryEngine(
    sql_database=sql_database,
    row_retriever=row_retriever,
    col_retriever=col_retriever,
    table_retriever=table_retriever,
)
```
- ›Adds cols_retrievers argument to NLSQLRetriever for column-level retrieval control.
- ›Adds row, col, and table retriever arguments to NLSQLTableQueryEngine for fine-grained SQL query engine configuration.
- ›Adds allow_parallel_tool_calls configurable argument to FunctionAgent.
- ›Adds search_filters_inference_schema client support to llama-index-indices-managed-llama-cloud.
- ›Adds stream_step and astream_step support to llama-index-agent-llm-compiler.
+22 moreshow less
- ›Overhauled BasicMCPClient in llama-index-tools-mcp to support all MCP features, including BasicMCPClient.with_oauth().
- ›Enhances SSE endpoint detection in llama-index-tools-mcp for broader MCP server compatibility.
- ›New llama-index-observability-otel [0.1.0] package adds OpenTelemetry integration for LlamaIndex observability.
- ›New llama-index-agent-azure-foundry [0.1.0] package adds Azure Foundry agent integration.
- ›New llama-index-llms-featherlessai [0.1.0] package adds Featherless AI LLM integration.
- ›New llama-index-llms-servam [0.1.1] package adds Servam AI LLM integration with an OpenAI-like interface.
- ›New llama-index-tools-brightdata [0.1.0] package adds Bright Data tool integration.
- ›Adds a simple embeddings cache implementation to llama-index-core.
- ›Adds Claude 4 model support to llama-index-llms-anthropic and llama-index-llms-bedrock-converse.
- ›Adds new OpenAI Responses API features (image generation, MCP call, code interpreter) to llama-index-llms-openai.
- ›Adds ctx context parameter support to BaseToolSpec functions with broader tool-calling overhauls.
- ›Adds async methods and blank index creation to llama-index-indices-managed-llama-cloud.
- ›Adds voyage-3.5 model support to llama-index-embeddings-voyageai.
- ›Adds retry configuration support to llama-index-embeddings-google-genai.
- ›Adds automatic context window detection to llama-index-llms-ollama.
- ›Adds default temperature support for Ollama models in llama-index-llms-ollama.
- ›Adds Vector Index Compression support to the Azure Cosmos DB Mongo vector store (llama-index-vector-stores-azurecosmosmongo).
- ›Adds filter support to check for the absence of a metadata key in llama-index-vector-stores-opensearch.
- ›Adds ability to create PostgresKVStore from an existing SQLAlchemy Engine in llama-index-storage-kvstore-postgres.
- ›Updates llama-index-postprocessor-rankllm-rerank to use the latest rank-llm SDK.
- ›Updates llama-index-tools-valyu to valyu 2.0.0.
- ›Updates llama-index-llms-cleanlab with new package name and updated models.
v0.12.37 May 20, 2025 · issue -366

LlamaIndex v0.12.37 adds Vectorize retriever and Desearch tool integrations, plus missing Bedrock client params.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.12.37 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.12.37
```
- ›Adds llama-index-retrievers-vectorize (v0.1.0) with a new Vectorize retriever integration.
- ›Adds llama-index-tools-desearch (v0.1.0) with a new Desearch tool integration.
- ›Adds missing client params for Bedrock Converse in llama-index-llms-bedrock-converse.
- ›Passes agent workflow kwargs into the start event in llama-index-core.
v0.12.35 May 8, 2025 · issue -366

LlamaIndex v0.12.35 adds memory revamp, Gel storage integrations, prefill tool kwargs, Anthropic citations, and new SlideNodeParser
└──▷ GET THIS VERSION
```
$ git clone --branch v0.12.35 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.12.35
```
- ›Adds prefilling partial tool kwargs support on FunctionTool, allowing callers to pre-bind arguments before the model completes the call.
- ›Adds indexed metadata fields to llama-index-vector-stores-postgres for faster filtered queries against document metadata.
- ›Adds FaissMapVectorStore to llama-index-vector-stores-faiss, providing a map-backed Faiss vector store variant.
- ›Introduces a memory revamp in llama-index-core with a new base class and prebuilt memory blocks for agent memory management.
- ›Adds four new Gel integrations at version 0.1.0: llama-index-storage-chat-store-gel, llama-index-storage-docstore-gel, llama-index-storage-kvstore-gel, and llama-index-storage-index-store-gel.
+7 moreshow less
- ›Adds llama-index-vector-stores-gel [0.1.0] as a new Gel-backed vector store integration.
- ›Adds SlideNodeParser integration in the new llama-index-node-parser-slide [0.1.0] package for parsing slide-format documents.
- ›Adds Anthropic citations and tool calls support to llama-index-llms-anthropic [0.6.12].
- ›Adds AutoEmbeddings integration from Chonkie in the new llama-index-embeddings-autoembeddings [0.1.0] package.
- ›Adds support for Meta Llama API as an LLM provider via llama-index-llms-meta [0.1.1].
- ›Adds Oxylabs readers in llama-index-readers-oxylabs [0.1.2] and llama-index-readers-web [0.4.1].
- ›Adds Cortex authentication enhancements to llama-index-llms-cortex [0.3.0].
v0.12.0 Nov 18, 2024 · issue -372

LlamaIndex v0.12.0 adds VLM support for NVIDIA, LlamaCloud file/ID APIs, Vectara custom prompts, and more.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.12.0 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.12.0
```
- ›Adds VLM (vision-language model) support to llama-index-multi-modal-llms-nvidia.
- ›Adds ID support for LlamaCloudIndex and new files endpoints in llama-index-indices-managed-llama-cloud.
- ›Adds option to skip waiting for ingestion when uploading files in llama-index-indices-managed-llama-cloud.
- ›Adds custom prompt parameter support to llama-index-indices-managed-vectara.
- ›Adds base URL extraction method to GithubRepositoryReader in llama-index-readers-github.
+3 moreshow less
- ›Allows passing additional kwargs to the Weaviate vector store in llama-index-vector-stores-weaviate.
- ›Allows passing custom params to the Confluence client in llama-index-readers-confluence.
- ›Adds dynamic triplet retrieval limit for KG/PG queries in llama-index-core.
└──▷ BREAKING ON UPGRADE
- !Python 3.8 is no longer supported; upgrading to v0.12.0 requires Python 3.9 or later.
- !Every llama-index-* package requires a version bump alongside llama-index-core 0.12.0 — mismatched package versions will break existing installs.
v0.10.68 Aug 21, 2024 · issue -375

LlamaIndex v0.10.68 adds nested workflow services, tool calling for Cohere/AI21, GigaChat LLM, and streaming token counts for OpenAI.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.10.68 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.10.68
```
└──▷ USE IT
Use the @step decorator without parentheses to register a workflow step with less boilerplate.
python
```
from llama_index.core.workflow import Workflow, step

class MyWorkflow(Workflow):
    @step
    async def my_step(self, ctx, ev):
        return ...
```
- ›Adds @step decorator support without parentheses in llama-index-core workflows, simplifying step registration syntax.
- ›Introduces workflow services (nested workflows) in llama-index-core, enabling workflows to be composed and reused as sub-components of larger workflows.
- ›Removes the requirement to specify the allowed_query_fields parameter when using cypher_validator in the TextToCypher retriever.
- ›Adds truncate support to llama-index-postprocessor-nvidia-rerank [0.2.1] and updates the default model to nvidia/nv-rerankqa-mistral-4b-v3.
- ›Adds streaming token count support to llama-index-llms-openai [0.1.31].
+10 moreshow less
- ›Adds tool calling support for achat in llama-index-llms-cohere [0.2.2].
- ›Adds AI21 Tools support to llama-index-llms-ai21 [0.3.2].
- ›Adds GigaChat LLM integration via new package llama-index-llms-gigachat [0.1.0].
- ›Adds token counting support for the Bedrock LLM integration in llama-index-llms-bedrock [0.1.13].
- ›Exposes structured schema for Amazon Neptune in llama-index-graph-stores-neptune [0.1.8].
- ›Adds static input shape support for OpenVINO embedding and reranker in llama-index-embeddings-openvino [0.2.1].
- ›Switches llama-index-embeddings-ollama [0.2.0] to use the native Ollama client for embeddings.
- ›Removes the OpenAI dependency from llama-index-core, reducing mandatory third-party coupling.
- ›Improves the llama-index-core token counter to handle more response types.
- ›Enhances the Google Drive reader in llama-index-readers-google [0.3.1] for improved functionality and usability.
v0.10.59 Aug 1, 2024 · issue -375

LlamaIndex v0.10.59 adds event-driven Workflows, LongRAG pack, FalkorDB graph store, GitLab reader, and function-calling for Ollama.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.10.59 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.10.59
```
└──▷ USE IT
Use the LongRAG pack to retrieve and answer over long documents with minimal chunking loss.
python
```
from llama_index.packs.longrag import LongRAGPack

pack = LongRAGPack(documents=documents, llm=llm)
response = pack.run("What are the key findings in this report?")
print(response)
```
- ›Introduces Workflow class in llama-index-core for event-driven orchestration of LlamaIndex pipelines.
- ›Adds llama-index-packs-longrag [0.1.0] — a new LlamaPack implementing the LongRAG retrieval pattern.
- ›Adds llama-index-graph-stores-falkordb [0.1.5] with FalkorDBPropertyGraphStore for property graph storage via FalkorDB.
- ›Adds llama-index-readers-gitlab [0.1.0] — a new GitLab reader integration for ingesting GitLab content.
- ›Adds llama-index-postprocessor-tei-rerank [0.1.0] — re-ranking support via Text Embedding Interface.
+8 moreshow less
- ›Adds llama-index-embeddings-textembed [0.0.1] — new embedding integration for the textembed backend.
- ›Adds function calling support and a toggle for it in llama-index-llms-ollama [0.2.2].
- ›Adds proper async embedding support to llama-index-embeddings-ollama [0.1.3].
- ›Adds HNSW index construction option to PGVectorStore in llama-index-vector-stores-postgres.
- ›Enhances MilvusVectorStore in llama-index-vector-stores-milvus with flexible index management for overwriting.
- ›Updates llama-index-llms-openllm to support OpenLLM 0.6.
- ›Expands span coverage for query pipeline tracing in llama-index-core.
- ›Adds feature to context chat engine allowing previous chunks to be inserted into the current context window.
v0.10.57 Jul 23, 2024 · issue -376

LlamaIndex v0.10.57 adds streaming tool-call extraction, KG property extraction, async BedrockConverse, and delete_nodes()/clear() across five vector stores.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.10.57 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.10.57
```
└──▷ USE IT
Filter vector context retrieval results to only those above a similarity threshold, reducing noisy context passed to the LLM.
python
```
from llama_index.core.retrievers import VectorContextRetriever

retriever = VectorContextRetriever(
    vector_store_index,
    similarity_score=0.75
)
nodes = retriever.retrieve("What is the access control policy?")
```
Purge all nodes from a Pinecone index (e.g., before a full re-ingestion) using the new clear() method.
python
```
from llama_index.vector_stores.pinecone import PineconeVectorStore

vector_store = PineconeVectorStore(pinecone_index=pinecone_index)
vector_store.clear()
```
- ›Adds optional similarity_score parameter to VectorContextRetriever to filter retrieved context by minimum similarity threshold.
- ›Adds property extraction (using property names and optional descriptions) for knowledge graphs in llama-index-core.
- ›Supports attaching output classes directly to LLMs for structured extraction.
- ›Adds streaming support for tool calling and structured extraction in llama-index-core.
- ›Implements delete_nodes() and clear() methods for Weaviate, OpenSearch, Milvus, Postgres, and Pinecone vector stores.
+3 moreshow less
- ›Implements async functionality in BedrockConverse (llama-index-llms-bedrock-converse v0.1.5).
- ›Enhances metadata filtering for MongoDB Atlas Vector Search in llama-index-vector-stores-mongodb.
- ›Updates Notion reader to handle duplicate pages and combined database+page IDs.
v0.10.52 Jul 3, 2024 · issue -376

LlamaIndex v0.10.52 adds Iceberg reader, MongoDB hybrid search, LiteLLM proxy embeddings, and async Azure AI Search methods.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.10.52 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.10.52
```
└──▷ USE IT
Use LiteLLM Proxy Server as your embeddings backend to route through a unified proxy endpoint.
python
```
from llama_index.embeddings.litellm import LiteLLMEmbedding

embed_model = LiteLLMEmbedding(model="text-embedding-ada-002", api_base="http://localhost:8000")
```
List all available Notion databases programmatically before loading data.
python
```
from llama_index.readers.notion import NotionPageReader

reader = NotionPageReader(integration_token="<token>")
databases = reader.list_databases()
print(databases)
```
- ›Adds list_databases method to llama-index-readers-notion for programmatic Notion database discovery.
- ›Adds llama-index-embeddings-litellm v0.1.0 integration supporting LiteLLM Proxy Server as an embeddings backend.
- ›Adds async methods to llama-index-vector-stores-azureaisearch for non-blocking Azure AI Search operations.
- ›Adds Hybrid Search and Full-Text Search to MongoDBAtlasVectorSearch in llama-index-vector-stores-mongodb.
- ›Adds llama-index-readers-iceberg v0.1.0 integration for reading Apache Iceberg tables into LlamaIndex.
+5 moreshow less
- ›Adds device selection (via sentence_transformers device choice) in llama-index-finetuning.
- ›Adds upstage tokenizer and token counting method to llama-index-llms-upstage.
- ›Adds API URL configuration to the Firecrawl reader in llama-index-readers-web.
- ›Adds automatic retry support to llama-index-readers-notion.
- ›Adds KDB.AI REST-compatible mode in llama-index-vector-stores-kdbai.
v0.10.42 May 31, 2024 · issue -378

LlamaIndex v0.10.42 adds NebulaGraph as a PropertyGraphStore backend and updates OpenLLM and PremAI SDK integrations.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.10.42 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.10.42
```
- ›Adds NebulaGraph support for PropertyGraphStore via the new llama-index-graph-stores-nebula 0.2.0 package, enabling NebulaGraph as a property graph backend.
- ›Updates llama-index-llms-openllm to support the OpenLLM 0.5 SDK.
- ›Updates llama-index-llms-premai for compatibility with the latest PremAI SDK.
v0.10.41 May 31, 2024 · issue -378

LlamaIndex v0.10.41 adds Mistral code and fill-in-middle models, embedding propagation to property graph retrievers, and streaming completion events.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.10.41 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.10.41
```
- ›Propagates embeddings from the index to the property graph retriever, enabling embedding-based graph retrieval without manual re-configuration.
- ›Adds the Mistral code model (llama-index-llms-mistralai 0.1.15) as a supported LLM integration.
- ›Adds fill-in-the-middle endpoint support for Mistral Codestral in llama-index-llms-mistralai.
- ›Adds missing instrumentation events for completion streaming in llama-index-core, enabling complete observability over streamed LLM responses.
- ›Uses the model kwarg for model name in the Gemini LLM integration (llama-index-llms-gemini 0.1.10).
+3 moreshow less
- ›Updates llama-index-llms-openllm to support OpenLLM 0.5 integrations.
- ›Adds safety setting support for the Vertex AI integration (llama-index-llms-vertex 0.1.8) to handle Pydantic errors.
- ›Adds support for path objects in the Smart PDF reader (llama-index-readers-smart-pdf-loader 0.1.5).
v0.10.40 May 29, 2024 · issue -378

LlamaIndex v0.10.40 adds PropertyGraphIndex, Neo4jPGStore, SecGPT integration, OCI Generative AI, and Hologres vector store support.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.10.40 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.10.40
```
- ›Adds PropertyGraphIndex to llama-index-core along with supporting abstractions for property graph-based indexing workflows.
- ›Adds Neo4jPGStore to llama-index-graph-stores-neo4j for property graph support backed by Neo4j.
- ›Adds llama-index-packs-secgpt [0.1.0] integrating SecGPT, a cybersecurity-focused LLM pack, into LlamaIndex.
- ›Adds llama-index-llms-oci-genai [0.1.0] and llama-index-embeddings-oci-genai [0.1.0] bringing Oracle Cloud Infrastructure (OCI) Generative AI support for both LLMs and embeddings.
- ›Adds llama-index-vector-stores-hologres [0.1.0] integrating the Hologres vector database as a new vector store backend.
+5 moreshow less
- ›Adds llama-index-indices-managed-dashscope [0.1.1] introducing a DashScope managed index.
- ›Adds support for Bedrock Titan Embeddings v2 in llama-index-embeddings-bedrock [0.2.0].
- ›Exposes the safe_serialization parameter from AutoModel in llama-index-embeddings-huggingface.
- ›Updates AutoPrevNextNodePostprocessor in llama-index-core to accept a custom response mode and LLM.
- ›Implements additional filter types for SimpleVectorStoreIndex in llama-index-core.
v0.10.35 May 7, 2024 · issue -378

LlamaIndex v0.10.35 adds NVIDIA NIM embeddings, LLM, and rerank support, plus new CRITIC/reflection agents and Vespa/Vertex AI vector stores.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.10.35 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.10.35
```
- ›Adds llama-index-llms-nvidia [0.1.0] with NVIDIA NIM LLM support via the new llama_index.llms.nvidia integration.
- ›Adds llama-index-embeddings-nvidia [0.1.0] with NVIDIA NIM embeddings support via the new llama_index.embeddings.nvidia integration.
- ›Adds llama-index-postprocessor-nvidia-rerank [0.1.0] with NVIDIA NIM rerank support.
- ›Adds llama-index-vector-stores-vespa [0.1.0] introducing a VectorStore integration for Vespa.
- ›Adds llama-index-vector-stores-vertexaivectorsearch [0.1.0] introducing Vertex AI Vector Search as a vector store backend.
+5 moreshow less
- ›Adds llama-index-agent-introspective [0.1.0] with CRITIC and reflection agent integrations.
- ›Adds encoding_type parameter to the JinaEmbedding class in llama-index-embeddings-jinaai.
- ›Updates MarkdownReader in llama-index-readers-file to parse text that appears before the first header.
- ›Adds Spider Web Loader to llama-index-readers-web.
- ›Expands instrumentation payloads in llama-index-core.
v0.10.34 May 3, 2024 · issue -378

LlamaIndex v0.10.34 adds structured planning agent, chat summary memory, hybrid retrieval, YouTube reader, and streaming expansions across multiple LLM integrations.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.10.34 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.10.34
```
- ›Adds ChatSummaryMemoryBuffer to llama-index-core for memory-efficient chat history management via summarization.
- ›Adds a structured planning agent to llama-index-core with an updated base class for planner agents.
- ›Updates HitRate and MRR retrieval metrics in llama-index-core to support Evaluation@K documents retrieved, and introduces RR (Reciprocal Rank) as a separate standalone metric.
- ›Adds hybrid retrieval mode to MilvusVectorStore in llama-index-vector-stores-milvus.
- ›Adds llama-index-vector-stores-firestore [0.1.0] — a new Firestore Vector Store integration.
+10 moreshow less
- ›Adds llama-index-readers-youtube-metadata [0.1.0] — a new YouTube Metadata Reader.
- ›Adds Browserbase Web Reader to llama-index-readers-web.
- ›Adds tool usage support to llama-index-llms-huggingface via the text-generation-inference integration.
- ›Adds streaming support to llama-index-llms-maritalk.
- ›Adds async support to llama-index-llms-ollama.
- ›Adds streaming support to llama-index-llms-nvidia-triton.
- ›Integrates mistral.rs as a new LLM backend in llama-index-llms-mistral-rs [0.1.0].
- ›Adds source_node.node_id verification matching to node parsers in llama-index-core.
- ›Allows ZillizCloudPipelineIndex to accept flexible parameters when creating pipelines.
- ›Excludes access control metadata keys from LLM and embedding calls in the SharePoint Reader.
v0.10.31 Apr 24, 2024 · issue -379

LlamaIndex v0.10.31 adds three new agents, two new readers, two new vector stores, and function-calling LLM programs
└──▷ GET THIS VERSION
```
$ git clone --branch v0.10.31 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.10.31
```
- ›Adds llama-index-agent-coa package (v0.1.0) with a new Chain-of-Abstraction (COA) agent integration.
- ›Adds llama-index-agent-lats package (v0.1.0) with an official LATS (Language Agent Tree Search) agent integration.
- ›Adds llama-index-agent-llm-compiler package (v0.1.0) with an LLMCompiler agent integration.
- ›Adds a function calling LLM program to llama-index-core.
- ›Adds llama-index-readers-openapi package (v0.1.0) with a reader for OpenAPI spec files.
+9 moreshow less
- ›Adds llama-index-vector-stores-awsdocdb package (v0.1.0) integrating AWS DocumentDB as a vector store backend.
- ›Adds streaming partial instances of Pydantic output class in OpenAIPydanticProgram via llama-index-program-openai.
- ›Adds support for passing custom headers to Anthropic LLM requests in llama-index-llms-anthropic.
- ›Adds Claude 3 Opus model support to the llama-index-llms-bedrock integration.
- ›Adds Llama 3 and Mixtral 8x22B model support to llama-index-llms-fireworks.
- ›Adds metadata filtering support to llama-index-vector-stores-neo4j.
- ›Adds index deletion functionality to WeaviateVectorStore in llama-index-vector-stores-weaviate.
- ›Updates IBM watsonx foundation models available in llama-index-llms-watsonx.
- ›Makes PydanticSingleSelector work with the async API in llama-index-core.
v0.10.30 Apr 17, 2024 · issue -379

LlamaIndex v0.10.30 adds LATS agent pack, two new embedding integrations, OR filter support, and intermediate QueryPipeline outputs.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.10.30 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.10.30
```
└──▷ USE IT
Filter vector store results using an OR condition to match documents from multiple sources.
python
```
from llama_index.core.vector_stores.types import MetadataFilters, MetadataFilter, FilterCondition

filters = MetadataFilters(
    filters=[
        MetadataFilter(key="source", value="arxiv"),
        MetadataFilter(key="source", value="pubmed"),
    ],
    condition=FilterCondition.OR,
)
results = index.as_retriever(filters=filters).retrieve("transformer models")
```
Use a token provider for Azure OpenAI embeddings so credentials refresh automatically before expiry.
python
```
from azure.identity import DefaultAzureCredential, get_bearer_token_provider
from llama_index.embeddings.azure_openai import AzureOpenAIEmbedding

token_provider = get_bearer_token_provider(
    DefaultAzureCredential(), "https://cognitiveservices.azure.com/.default"
)
embed_model = AzureOpenAIEmbedding(
    model="text-embedding-ada-002",
    deployment_name="my-deployment",
    azure_endpoint="https://<your-resource>.openai.azure.com/",
    azure_ad_token_provider=token_provider,
)
```
- ›Adds OR filter condition support to the simple vector store, enabling more flexible metadata filtering alongside existing AND conditions.
- ›Exposes azure_ad_token_provider argument in both llama-index-embeddings-azure-openai and llama-index-llms-azure-openai to support token expiration/refresh scenarios.
- ›Adds httpx_async_client option to llama-index-embeddings-cohere for async HTTP client customization.
- ›New llama-index-embeddings-ipex-llm integration (v0.1.0) adds embedding support via Intel IPEX-LLM.
- ›New llama-index-embeddings-octoai integration (v0.1.0) adds embedding support via OctoAI.
+7 moreshow less
- ›Adds support for loading 'low-bit format' models in the IpexLLM LLM integration.
- ›Adds support for the open-mixtral-8x22b model in llama-index-llms-mistralai.
- ›New llama-index-packs-agents-lats (v0.1.0) introduces the LATS (Language Agent Tree Search) agent pack.
- ›New llama-index-readers-web Firecrawl Web Loader adds web crawling/loading via Firecrawl.
- ›New llama-index-vector-stores-vearch integration (v0.1.0) adds Vearch as a supported vector store.
- ›Adds intermediate outputs to QueryPipeline, enabling inspection of pipeline step results.
- ›Switches llama-index-vector-stores-milvus to batch insertions for improved write throughput.
v0.10.29 Apr 14, 2024 · issue -379

LlamaIndex v0.10.29 adds OpenVINO LLMs and reranking, Couchbase and Bedrock vector/retrieval integrations, Chain-of-Abstraction agent pack, and Mistral Large on Bedrock.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.10.29 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.10.29
```
- ›Adds llama-index-llms-openvino (0.1.0) — new OpenVino LLM integration installable via pip install llama-index-llms-openvino.
- ›Adds llama-index-postprocessor-openvino-rerank OpenVINO reranking postprocessor support.
- ›Adds llama-index-retrievers-bedrock (0.1.0) — Amazon Bedrock knowledge base integration as a retriever.
- ›Adds llama-index-retrievers-mongodb-atlas-bm25-retriever (0.1.3) — MongoDB Atlas BM25 retriever.
- ›Adds llama-index-vector-stores-couchbase (0.1.0) — Couchbase as a vector store.
+7 moreshow less
- ›Adds llama-index-packs-agents-coa (0.1.0) — Chain-of-Abstraction agent pack.
- ›Adds Mistral Large model support in llama-index-llms-bedrock.
- ›Enables choice of either Predibase-hosted or HuggingFace-hosted fine-tuned adapters in the llama-index-llms-predibase integration.
- ›Modernizes llama-index-vector-stores-redis (0.2.0) to use redisvl.
- ›Adds metadata field retrieval support in llama-index-vector-stores-milvus.
- ›Updates llama-index-llms-predibase to the latest Predibase API.
- ›Modernizes GuardrailsOutputParser in llama-index-output-parsers-guardrails.
└──▷ BREAKING ON UPGRADE
- !PandasQueryEngine and PandasInstruction parser are moved out of llama-index-core into llama-index-experimental; existing code will break until updated with pip install -U llama-index-experimental and the new import from llama_index.experimental.query_engine import PandasQueryEngine.
v0.10.28 Apr 9, 2024 · issue -379

LlamaIndex v0.10.28 adds Anthropic tool calling, OpenVINO embeddings, ipex-llm integration, and multilingual Wikipedia support.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.10.28 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.10.28
```
└──▷ USE IT
Return a tool's output directly to the user without further LLM synthesis — useful for lookup tools where the raw result is the final answer.
python
```
from llama_index.core.tools import FunctionTool

def lookup_price(ticker: str) -> str:
    return f"${ticker}: 142.00"

price_tool = FunctionTool.from_defaults(
    fn=lookup_price,
    return_direct=True,
)
```
Fetch multilingual Wikipedia articles for ingestion — useful for building RAG pipelines over non-English content.
python
```
from llama_index.readers.wikipedia import WikipediaReader

reader = WikipediaReader()
docs = reader.load_data(pages=["Louvre"], lang="fr")
```
- ›Adds return_direct option to tool metadata in llama-index-core, letting tools short-circuit the agent loop and return their output directly to the caller.
- ›Adds async_postprocess_nodes to the RankGPT postprocessor in llama-index-core, enabling fully async reranking pipelines.
- ›Adds thread-safe and coroutine-safe instrumentation spans in llama-index-core, making telemetry safe for concurrent and async workloads.
- ›Adds in-memory loading for non-default filesystems in PDFReader (llama-index-core), enabling PDF ingestion from remote or custom storage backends.
- ›Adds SynthesizeComponent to shortcut imports in llama-index-core.
+9 moreshow less
- ›Adds streaming support for DenseXRetrievalPack in llama-index-packs-dense-x-retrieval.
- ›Adds retry logic to the batch eval runner in llama-index-core, improving resilience of bulk evaluation jobs.
- ›Adds output parser passthrough to the guideline evaluator in llama-index-core.
- ›Adds support for indented code block fences in the markdown node parser in llama-index-core.
- ›Introduces llama-index-embeddings-openvino v0.1.5 with initial support for OpenVINO-accelerated embeddings.
- ›Adds Anthropic tool calling support in llama-index-llms-anthropic v0.1.9.
- ›Introduces llama-index-llms-ipex-llm v0.1.1 with ipex-llm LLM integration and support for multiple data types.
- ›Adds multilingual support to the Wikipedia reader in llama-index-readers-wikipedia.
- ›Adds metadata field retrieval from Milvus in llama-index-vector-stores-milvus.
v0.10.27 Apr 4, 2024 · issue -379

LlamaIndex v0.10.27 adds Databricks, Cloudflare Workers AI, and Neptune Analytics integrations alongside Cohere Command R+ and RankGPT support.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.10.27 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.10.27
```
- ›Adds span_id attribute to Events in the instrumentation layer (llama-index-core).
- ›Adds node-postprocessors support to retriever_tool (llama-index-core).
- ›Adds FLAREInstructQueryEngine delegation to the retriever API when the query engine supports it (llama-index-core).
- ›New llama-index-llms-databricks [0.1.0] integration with the Databricks LLM API.
- ›New llama-index-embeddings-cloudflar-workersai [0.1.0] text embedding integration with Cloudflare Workers AI.
+8 moreshow less
- ›New llama-index-vector-stores-neptune [0.1.0] adds Neptune Analytics as a vector store backend.
- ›Adds support for the Cohere Command R+ model in llama-index-llms-cohere.
- ›Adds RankGPT support inside RankLLM via llama-index-postprocessor-rankllm-rerank.
- ›Adds ability to pass custom HTTP headers to the Anthropic client in llama-index-llms-anthropic.
- ›Adds support for loading CLIP models from a local file path in llama-index-embeddings-clip.
- ›Updates Watsonx foundation models and base model names in llama-index-llms-watsonx.
- ›Changes llama-index-readers-microsoft-sharepoint to use a recursive reading strategy by default.
- ›Replaces the Redis driver with the FalkorDB driver in llama-index-graph-stores-falkordb.
└──▷ BREAKING ON UPGRADE
- !The llama-index-graph-stores-falkordb package now uses the FalkorDB driver instead of the Redis driver; any setup relying on the Redis driver will break on upgrade.
- !The llama-index-readers-microsoft-sharepoint package now uses the recursive strategy by default, which may change the set of documents retrieved for existing SharePoint configurations.
v0.10.19 Mar 12, 2024 · issue -380

LlamaIndex v0.10.19 adds log-probability support, labelled datasets, SQL table comments, nested metadata filters, and new LLM model support.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.10.19 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.10.19
```
- ›Adds LogProb type to the ChatResponse object in llama-index-core, exposing token-level log probabilities from model responses.
- ›Adds table comments to SQL table schemas in SQLDatabase in llama-index-core, giving the query engine richer schema context.
- ›Introduces LabelledSimpleDataset in llama-index-core for working with labelled training/evaluation data.
- ›Adds support for nested metadata filters in llama-index-vector-stores-postgres.
- ›Adds support for the command-r model in llama-index-llms-cohere.
+2 moreshow less
- ›Adds support for latest and open models in llama-index-llms-mistralai.
- ›Introduces automatic retries for rate limits in the OpenAI LLM class in llama-index-core.
v0.10.17 Mar 7, 2024 · issue -380

LlamaIndex v0.10.17 adds relative/dist-based fusion scoring, Anthropic multimodal models, a finance chat llama-pack, and SQL refine templates.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.10.17 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.10.17
```
└──▷ USE IT
Use distance-based score normalization in a fusion retriever to improve ranking across heterogeneous retrievers.
python
```
from llama_index.core.retrievers import QueryFusionRetriever

retriever = QueryFusionRetriever(
    retrievers=[retriever_a, retriever_b],
    mode="dist_based_score",
    num_queries=4,
)
nodes = retriever.retrieve("What is the capital of France?")
```
- ›Adds relative_score and dist_based_score scoring modes to QueryFusionRetriever in llama-index-core.
- ›Adds support for a refine template in BaseSQLTableQueryEngine via llama-index-core.
- ›Adds support for Anthropic multimodal models haiku and sonnet in llama-index-multi-modal-llms-anthropic.
- ›Adds new llama-index-packs-finchat llama-pack for hierarchical agents combined with finance chat workflows.
- ›Inherits metadata to summaries in DocumentSummaryIndex in llama-index-core.
v0.10.14 Feb 28, 2024 · issue -381

LlamaIndex v0.10.14 adds llama-index-networks, Jina reranker, Brave/DuckDuckGo agent search tools, Friendli LLM, and ChromaDB metadata-only queries.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.10.14 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.10.14
```
- ›Adds llama-index-networks package enabling federated/networked index querying across distributed LlamaIndex deployments.
- ›Adds Jina reranker integration for post-retrieval result reranking.
- ›Adds DuckDuckGo agent search tool for use with LlamaIndex agents.
- ›Adds Brave Search tool for use with LlamaIndex agents.
- ›Adds Friendli LLM integration as a new supported language model provider.
+2 moreshow less
- ›Adds metadata-only query support for ChromaDB vector store, enabling lightweight filtering without full vector retrieval.
- ›Adds helper functions for ChatML format handling.
v0.10.13 Feb 26, 2024 · issue -381

LlamaIndex v0.10.13 adds fsspec support, mistral-large, last-token pooling for HuggingFace embeddings, and a KodaRetriever pack.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.10.13 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.10.13
```
- ›Adds fsspec support to SimpleDirectoryReader, enabling reads from any fsspec-compatible filesystem (S3, GCS, ADLS, etc.).
- ›Adds a llama-pack for KodaRetriever with on-the-fly alpha tuning for hybrid retrieval weighting.
- ›Supports mistral-large as a new model option.
- ›Adds last-token pooling mode for HuggingFace embedding models such as SFR-Embedding-Mistral.
v0.10.7 Feb 19, 2024 · issue -381

LlamaIndex v0.10.7 adds a Self-Discover LlamaPack for structured reasoning workflows.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.10.7 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.10.7
```
- ›Adds Self-Discover LlamaPack, enabling structured self-discovery reasoning workflows via the llamapack interface.
v0.10.6 Feb 18, 2024 · issue -381

LlamaIndex v0.10.6 adds NomicHFEmbedding and MinioReader integrations.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.10.6 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.10.6
```
- ›Adds NomicHFEmbedding class for Nomic embedding model support via Hugging Face.
- ›Adds MinioReader class for ingesting data directly from MinIO object storage.
v0.10.1 Feb 12, 2024 · issue -381

LlamaIndex v0.10 splits into a llama-index-core package plus hundreds of separate integration packages, and deprecates ServiceContext.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.10.1 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.10.1
```
└──▷ USE IT
Use a namespace import that still works after the package split, without changing existing code.
python
```
from llama_index.llms.openai import OpenAI

llm = OpenAI(model="gpt-4")
```
- ›Introduces llama-index-core as a standalone PyPI package, with all integrations (LLMs, embeddings, vector stores, data loaders, callbacks, agent tools) split into individually versioned PyPI packages while preserving namespace imports (e.g. from llama_index.llms.openai import OpenAI still works).
- ›Consolidates the former llama-hub repository into the main llama_index repo under llama-index-integrations, making LlamaHub the single registry for all integrations.
- ›Deprecates ServiceContext in favour of directly specifying arguments or setting a global default, removing the centralized abstraction for managing LLMs, embeddings, chunk sizes, and callbacks.
└──▷ BREAKING ON UPGRADE
- !Integrations are no longer bundled in the monolithic llama_index package; existing code that imports integration classes may break until the corresponding separate integration package is installed.
- !ServiceContext is deprecated — code that constructs or passes a ServiceContext object will need to be migrated to direct argument passing or global defaults.
v0.9.16 Dec 18, 2023 · issue -383

LlamaIndex v0.9.16 adds step-wise agent execution, OpenRouter integration, Neo4j hybrid search, and Google service account auth.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.9.16 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.9.16
```
- ›Adds step-wise (pause-and-resume) agent execution via the agent refactor, enabling finer-grained control over multi-step reasoning loops.
- ›Adds OpenRouter as a supported LLM provider, with a Mixtral demo included.
- ›Adds hybrid search support to the Neo4j vector store.
- ›Adds support for auth service accounts for Google Semantic Retriever.
v0.9.12 Dec 5, 2023 · issue -383

LlamaIndex v0.9.12 adds vLLM support, Python 3.12 compatibility, and claude-2.1 model name alongside a new async client option.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.9.12 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.9.12
```
- ›Adds reuse_client option to OpenAI/Azure integrations — set to False to reduce async timeout errors.
- ›Adds support for vLLM as an LLM backend.
- ›Adds support for the claude-2.1 model name in the Anthropic integration.
- ›Adds support for Python 3.12.
v0.9.10 Nov 30, 2023 · issue -384

LlamaIndex v0.9.10 adds advanced metadata filtering, new Bedrock embedding models, PromptLayer callbacks, and OpenAI Assistant file-ID reuse.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.9.10 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.9.10
```
- ›Adds advanced metadata filters for vector stores, replacing the deprecated ExactMatchFilter with a more capable filtering API.
- ›Adds new Amazon Bedrock embedding models via the existing Bedrock Embeddings integration.
- ›Adds PromptLayer callback integration for logging and observability of LLM prompt/response chains.
- ›Enables reuse of existing file IDs in OpenAIAssistant, avoiding redundant file uploads on repeated runs.
└──▷ BREAKING ON UPGRADE
- !ExactMatchFilter is deprecated in favour of the new advanced metadata filter API; usages of ExactMatchFilter should be migrated.
v0.9.9 Nov 29, 2023 · issue -384

LlamaIndex v0.9.9 adds LlamaDataset abstractions, metadata filtering, and MMR mode for AstraDB vector store.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.9.9 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.9.9
```
- ›Adds metadata filtering and MMR (Maximal Marginal Relevance) mode support to AstraDBVectorStore.
- ›Introduces new abstractions for LlamaDataset in the evaluation module.
- ›Allows latest scikit-learn versions as a compatible dependency.
└──▷ BREAKING ON UPGRADE
- !QueryResponseDataset and DatasetGenerator in the evaluation module are deprecated and began their deprecation cycle in v0.9.9.
- !LocalAI integration began its deprecation cycle in v0.9.9.
v0.9.8 Nov 26, 2023 · issue -384

LlamaIndex v0.9.8 adds async metadata extraction, ObjectIndex persistence, and character-index tracking in nodes.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.9.8 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.9.8
```
- ›Adds persist and persist_from_dir methods to ObjectIndex for saving and restoring index state to disk.
- ›Adds async metadata extraction with pipeline support for non-blocking ingestion workflows.
v0.5.10 Apr 7, 2023 · issue -391

LlamaIndex 0.5.10 adds hybrid sparse-dense search, Milvus integration, and in-memory Qdrant support.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.5.10 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.5.10
```
- ›Adds sparse-dense hybrid search support for Pinecone and Weaviate vector stores.
- ›Adds Milvus vector store integration.
- ›Adds in-memory Qdrant vector store support.
v0.5.0 Mar 28, 2023 · issue -392

LlamaIndex v0.5.0 overhauls its data model, composability API, and query pipeline with a new migration tool.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.5.0 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.5.0
```
└──▷ TRY IT
Migrate a saved tree index JSON file from v0.4.x to v0.5.0 before loading it in the new version.
```
$ python gpt_index/tools/migrate_v1_to_v2.py --v1_path tree_v1.json --index_struct_type tree --v2_path tree_v2.json
```
Compose a hierarchical graph from multiple sub-indices, each with a summary that guides top-level routing.
python
```
from llama_index import ComposableGraph

graph = ComposableGraph.build_from_indices(
    [index_a, index_b],
    summaries=["Summary of index A", "Summary of index B"]
)
```
- ›Adds from_documents class method on index classes as the new entry point for feeding documents directly into an index.
- ›Introduces ServiceContext container to consolidate custom LLMs, embedding models, chunk sizes, and prompt helpers into a single argument.
- ›Introduces ComposableGraph.build_from_indices(subindices, summaries) as the new API for composing hierarchical index graphs, backed by a CompositeIndexStruct.
- ›Adds index.index_struct.index_id and index.index_struct.summary as the canonical fields for setting index identity and summary metadata.
- ›Adds a migration tool at gpt_index/tools/migrate_v1_to_v2.py with --v1_path, --index_struct_type, and --v2_path flags to upgrade saved index JSON from 0.4.x to 0.5.0.
+3 moreshow less
- ›Introduces retrieve and synthesize methods on query classes to decouple node selection from answer synthesis.
- ›Nodes are now stored in DocumentStore instead of IndexStruct, enabling reuse of the same node across multiple indices without duplication.
- ›Node data model now tracks relationships between document chunks (e.g. ordering, source document) independently of any index struct.
└──▷ BREAKING ON UPGRADE
- !Index constructors now accept Node objects instead of Document objects; use the new from_documents class method to retain the previous document-based API.
- !index.set_doc_id is removed; set the index ID via index.index_struct.index_id = <value> instead.
- !The composable graph API has changed; replace previous graph construction calls with ComposableGraph.build_from_indices.
- !Common constructor arguments (LLM, embedding model, chunk size, prompt helper) must now be passed via a ServiceContext container rather than directly.
- !Saved index JSON files from 0.4.x are not directly compatible with 0.5.0 and must be migrated using gpt_index/tools/migrate_v1_to_v2.py.
v0.4.36 Mar 23, 2023 · issue -392

LlamaIndex 0.4.36 adds async support for composed graphs and Pinecone index sharing across multiple vector indices.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.4.36 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.4.36
```
- ›Expands async functionality to work over composed graphs, enabling concurrent operations across multi-index pipelines.
- ›Expands Pinecone integration to allow a single Pinecone index to be shared among multiple vector indices, enabling use on the free plan.
v0.4.28 Mar 14, 2023 · issue -392

LlamaIndex v0.4.28 adds native image ingestion and querying with image sources in results.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.4.28 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.4.28
```
- ›Adds a native image Document format, enabling image data to be ingested directly into LlamaIndex pipelines.
- ›Supports text-based queries over ingested image documents.
- ›Returns image sources alongside query results, surfacing which images contributed to a response.
v0.4.26 Mar 11, 2023 · issue -392

LlamaIndex 0.4.26 adds a Node postprocessor abstraction for post-retrieval filtering and re-ranking.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.4.26 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.4.26
```
- ›Adds a 'Node' postprocessor abstraction enabling additional filtering and retrieval operations on top of retrieved documents.
v0.4.25 Mar 10, 2023 · issue -392

LlamaIndex 0.4.25 adds a token-usage optimizer and Document sync support for indexes.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.4.25 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.4.25
```
- ›Adds an optimizer that can reduce LLM token usage by up to 50% or more.
- ›Enables syncing Document updates with an existing index.
v0.4.23 Mar 8, 2023 · issue -392

LlamaIndex 0.4.23 adds an empty Index type, upgraded LangChain agent memory integration, and a Steamship file reader.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.4.23 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.4.23
```
- ›Adds a Steamship file reader integration for ingesting files from Steamship into the index.
- ›Adds an 'empty' Index type to explicitly combine prior LLM knowledge with a knowledge corpus.
- ›Upgrades the LlamaIndex memory module integration with LangChain agents.
v0.4.21 Mar 6, 2023 · issue -392

LlamaIndex v0.4.21 adds a new JSON reader and improved ChatGPT integrations.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.4.21 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.4.21
```
- ›Adds a new JSON reader with novel JSON parsing, available in both LlamaIndex and LlamaHub.
- ›Improves ChatGPT integrations.
v0.4.13 Feb 25, 2023 · issue -393

LlamaIndex 0.4.13 adds embedding-based KG Index queries and multi-file LlamaHub loader support via download_loader.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.4.13 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.4.13
```
- ›Extends download_loader integration with LlamaHub to support complex loaders that require multiple files, enabling integrations such as the GitHub loader.
- ›Enables embedding-based querying of the Knowledge Graph (KG) Index as an alternative to exact keyword matching.
v0.4.12 Feb 24, 2023 · issue -393

LlamaIndex v0.4.12 lets you pass a nested index as table context to the SQL index, tackling large schema prompts.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.4.12 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.4.12
```
- ›Adds support for passing table context — including another index as the context source — to the SQL index, enabling text-to-SQL over databases with too many tables and columns to fit in a single prompt.
v0.4.11 Feb 24, 2023 · issue -393

LlamaIndex 0.4.11 adds async vector index construction and decouples vector storage from index logic.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.4.11 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.4.11
```
- ›Adds async support to vector index construction, enabling non-blocking index builds.
- ›Decouples vector storage from index build and query logic, laying the groundwork for new vector store integrations.
v0.4.8 Feb 21, 2023 · issue -393

LlamaIndex v0.4.8 adds customizable text splitters per index and a use_gpt_index_import option for LlamaHub loaders.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.4.8 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.4.8
```
└──▷ USE IT
Retain legacy gpt_index imports in LlamaHub loaders while migrating to llama_index at your own pace.
python
```
from llama_index import download_loader

SimpleWebPageReader = download_loader('SimpleWebPageReader', use_gpt_index_import=True)
```
- ›Adds use_gpt_index_import option to download_loader — set to True to retain gpt_index imports when LlamaHub loaders now default to llama_index.
- ›Adds ability to customize the text splitter for a given index.
└──▷ BREAKING ON UPGRADE
- !All LlamaHub loaders now import from llama_index instead of gpt_index by default; code relying on gpt_index imports from download_loader will break unless use_gpt_index_import=True is set.
v0.4.7 Feb 20, 2023 · issue -393

LlamaIndex v0.4.7 adds a Playground module for comparing indexes, models, and embeddings side by side.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.4.7 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.4.7
```
- ›New Playground module lets practitioners test multiple indexes, models, and embeddings simultaneously and compare results in one place.
v0.4.6 Feb 19, 2023 · issue -393

LlamaIndex v0.4.6 adds async tree_summarize queries and embedding batching for 3-5x faster responses and faster vector index construction.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.4.6 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.4.6
```
- ›Adds async support for tree_summarize queries, delivering 3-5x faster query responses.
- ›Adds embedding batching to accelerate vector index construction.
v0.4.5 Feb 18, 2023 · issue -393

LlamaIndex 0.4.5 adds KG index triplet tracking in sources and reduces index JSON file size.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.4.5 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.4.5
```
- ›Tracks triplets for the Knowledge Graph (KG) index in sources, making graph relationships visible in query provenance.
- ›Significantly reduces index JSON file size by removing unnecessary information.
v0.4.4 Feb 16, 2023 · issue -393

LlamaIndex v0.4.4 adds QueryBundle and QueryTransform abstractions for finer control over query embedding and transformation.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.4.4 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.4.4
```
- ›Adds QueryBundle abstraction to separate the query string from the string used for embedding lookup, enabling independent control of retrieval vs. generation queries.
- ›Adds QueryTransform class to transform queries within data structures, with HyDE (Hypothetical Document Embeddings) as the first implementation.
v0.4.3 Feb 14, 2023 · issue -393

LlamaIndex v0.4.3 adds a Knowledge Graph index for triplet extraction and query-time KG traversal.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.4.3 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.4.3
```
- ›Adds a Knowledge Graph index that builds a KG by extracting triplets from documents and leverages it at query time.
v0.4.2 Feb 13, 2023 · issue -393

GPT Index 0.4.2 adds caching to download_loader and exposes Pinecone kwargs across all index operations.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.4.2 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.4.2
```
- ›Adds caching to download_loader so loaders are served from local cache instead of re-downloading from llamahub.ai on every call.
- ›Exposes Pinecone kwargs on all index operations for the Pinecone index, enabling fine-grained control over Pinecone API calls.
v0.4.1 Feb 10, 2023 · issue -393

LlamaIndex v0.4.1 adds an Azure OpenAI example notebook, a GitHub repository loader, and an updated OpenAI retry policy.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.4.1 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.4.1
```
- ›Adds a GitHub repository loader for ingesting code and content directly from GitHub repositories.
- ›Adds an Azure OpenAI example notebook demonstrating integration with Azure-hosted OpenAI endpoints.
- ›Updates the retry policy for OpenAI API calls.
v0.4.0 Feb 8, 2023 · issue -393

LlamaIndex v0.4.0 replaces print statements with full Python logger support throughout the codebase.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.4.0 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.4.0
```
- ›Adds full Python logging module support throughout the codebase, replacing all print statements and enabling standard log routing, filtering, and formatting.
└──▷ BREAKING ON UPGRADE
- !The verbose parameter has been removed from all APIs; configure output verbosity using Python's standard logging module instead.
v0.3.6 Feb 7, 2023 · issue -393

LlamaIndex v0.3.6 adds an mbox parser so email archives can be fed directly into an index.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.3.6 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.3.6
```
- ›Adds a parser/reader for .mbox files, enabling email archives to be ingested as index documents.
v0.3.5 Feb 5, 2023 · issue -393

LlamaIndex v0.3.5 adds save/load from string for indices and graphs, and drops required query_configs for recursive queries.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.3.5 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.3.5
```
- ›Adds save/load from string for indices and graphs, enabling persistence to sources beyond disk.
- ›Removes the requirement to specify query_configs for recursive queries — default configs are used automatically.
v0.3.2 Feb 1, 2023 · issue -393

LlamaIndex v0.3.2 adds Qdrant as both a data source reader and a vector index store.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.3.2 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.3.2
```
- ›Adds Qdrant integration, enabling Qdrant to be used both as a data reader (source) and as a vector store for your index.
v0.3.1 Jan 31, 2023 · issue -394

LlamaIndex v0.3.1 adds ObsidianReader for parsing Markdown vaults with automatic hyperlink and image removal.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.3.1 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.3.1
```
- ›Adds ObsidianReader to parse a directory of Markdown files (Obsidian vaults or any Markdown collection), automatically stripping hyperlinks and images.
v0.3.0 Jan 30, 2023 · issue -394

LlamaIndex v0.3.0 improves the composability interface for building composite indices.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.3.0 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.3.0
```
- ›Improves the composability interface for combining multiple indices into composite query structures.
└──▷ BREAKING ON UPGRADE
- !The composability interface has changed; existing code using composable indices must be updated — see the ComposableIndices notebook for migration details.
v0.2.17 Jan 29, 2023 · issue -394

LlamaIndex v0.2.17 adds file-level control to SimpleDirectoryReader and recursive Notion child-page reading.
└──▷ GET THIS VERSION
```
$ git clone --branch v0.2.17 https://github.com/run-llama/llama_index.git
# already have the repo? check out this version:
$ git checkout v0.2.17
```
└──▷ USE IT
Load only a specific subset of files from a mixed directory — useful in CI pipelines where you want to index only changed documents.
python
```
from llama_index import SimpleDirectoryReader

reader = SimpleDirectoryReader(file_paths=['docs/overview.pdf', 'docs/changelog.md'])
documents = reader.load_data()
```
- ›Adds file_paths argument to SimpleDirectoryReader, letting callers specify an explicit list of files instead of an entire directory.
- ›Upgrades the Notion reader to recursively read child pages in addition to top-level pages.

LlamaIndex

Summary

What LlamaIndex answers

Examples

Command line

Release history