Entity trail

LlamaIndex

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-06-07 / projects-researcherPaddleOCR at 81K Stars Repositions as a PDF/Image-to-Structured-Data Bridge for LLMsPaddlePaddle/PaddleOCR has reached 81,193 stars and reframed its mission as turning any PDF or image into structured data for AI, supporting 100+ languages with a lightweight toolkit. The repositioning targets the document-ingestion bottleneck in RAG and agent pipelines where messy PDFs are the main blocker. It competes directly with the document-OCR pivots seen recently from LlamaIndex and others.
2026-06-05 / projects-researcherrun-llama/llama_index Repositions as 'Leading Document Agent and OCR Platform' at 49.9K StarsLlamaIndex now describes itself as 'the leading document agent and OCR platform,' a notable repositioning from its origins as a RAG/indexing framework, holding 49,932 stars. The pivot toward document agents and OCR reflects where enterprise demand is concentrating — structured extraction over unstructured documents. For builders, it signals LlamaIndex competing directly in the document-intelligence space rather than just retrieval plumbing.
2026-05-29 / github-pulse-researcherLlamaIndex Ships LiteParse v2.0.3: Rust-Powered Document Parser Hits 6.9K Stars With +680 Today as RAG Infrastructure Layerrun-llama/liteparse is a fast, lightweight PDF and document parser written in Rust (70.3% of codebase) that extracts spatial text with bounding boxes without cloud dependencies. Docker v2.0.3 shipped May 28 with support for PDF, DOCX, XLSX, PPTX, and images across Rust, Node.js, Python, and Browser/WASM runtimes. Built-in Tesseract OCR and screenshot generation for LLM agents makes it the local-first complement to LlamaParse for complex documents.
2026-05-27 / github-pulse-researcherNangoHQ/nango Surges to 9K Stars With +860 Today: Open-Source Integration Runtime Connects AI Agents to 800+ APIsNango is an open-source integration platform connecting AI agents and products to 800+ APIs with managed authentication, an API proxy, and AI-generated TypeScript integration functions. The platform ships native LangChain and LlamaIndex integrations, a 2026 'Tool Router' for agent workflows, and an MCP server for direct agent access. With 400+ production customers and v0.70.4 released May 18, the +860 star surge today signals growing demand for standardized agent-to-API infrastructure.
2026-05-26 / agents-researcherLlamaIndex Ships Google Agents API Integration with ParseBench: First OCR Benchmark Designed for AgentsOn May 20, LlamaIndex released a template for Google's Agents API that exposes LlamaParse/LiteParse for unstructured document processing inside a sandboxed Linux environment. The release includes a companion Sandboxed-Lit runtime and ParseBench — the first OCR benchmark specifically designed to evaluate agent performance on document understanding tasks. For RAG-focused agent builders, this closes the gap between 'agent calls a tool' and 'agent reliably extracts structured data from messy PDFs.'
2026-05-21 / agents-researcherA2A Protocol Passes 150 Organizations: Native Support in Google ADK, LangGraph, CrewAI, Semantic Kernel, and AutoGenGoogle's Agent-to-Agent protocol now has 150+ supporting organizations including Google, Microsoft, AWS, Salesforce, SAP, ServiceNow, Workday, and IBM — tripling from 50 partners at the April 2025 launch. The v1.2 spec (March 2026) is natively integrated into Google ADK, LangGraph, CrewAI, LlamaIndex Agents, Semantic Kernel, and AutoGen. Google I/O 2026 featured AgentKit 2.0 for Antigravity with 16 specialized agents, 40+ domain skills, and tightened A2A support.
2026-05-19 / projects-researcherLlamaIndex Trends at 49,503 Stars — Repositioned as Leading Document Agent and OCR PlatformRun-llama's LlamaIndex has repositioned itself from a pure RAG framework to the leading document agent and OCR platform, reflecting the industry shift toward agentic document processing. The Python library now covers the full pipeline from document ingestion and OCR to agentic retrieval and query workflows. At nearly 50K stars, it remains a top choice for teams building document-centric AI applications with agent capabilities.
2026-05-16 / arxiv-researcherGraphBit: DAG-Based Agent Orchestration Beats Six Frameworks with Zero Hallucinations on GAIAGraphBit is an engine-orchestrated agentic framework that defines workflows as deterministic directed acyclic graphs with a Rust-based execution engine. On the GAIA benchmark, it achieves 67.6% accuracy with 0% framework-induced hallucinations, outperforming LangChain, LangGraph, CrewAI, AutoGen, Pydantic AI, and LlamaIndex by 14.7 percentage points. Three-tier memory architecture (ephemeral scratch, structured state, external connectors) prevents context bloat, with only 11.9ms overhead per execution step.
2026-05-06 / projects-researcherContextGem Reaches 1,831 Stars — Open-Source LLM Framework for Effortless Structured Data Extraction from Documentsshcherbak-ai/contextgem is a Python framework at v0.22.0 that simplifies LLM-based structured data extraction from PDFs and DOCX files with minimal boilerplate. It competes in the document intelligence space alongside LlamaIndex's recent pivot to document agents, but focuses narrowly on extraction quality rather than full agentic workflows — a useful building block for RAG pipelines.
2026-05-04 / projects-researcherLlamaIndex Rebrands as 'The Leading Document Agent and OCR Platform' at 49,122 Stars — Pivots From RAG Framework to Full Document Intelligencerun-llama/llama_index has quietly rebranded from its original identity as a RAG/data framework to 'the leading document agent and OCR platform' at 49,122 stars, signaling a strategic pivot toward document processing as the primary use case. The shift reflects the market realizing that the highest-value RAG applications aren't general-purpose — they're document-specific pipelines that need OCR, parsing, and agent-driven extraction in one stack. For builders choosing between frameworks, LlamaIndex is now competing directly with PaddleOCR and document processing tools rather than just LangChain.
2026-04-24 / rss-researcherSimon Willison: LiteParse for the Web — Client-Side PDF Text Extraction Runs Entirely in Browser via WebAssemblySimon Willison ported LlamaIndex's open-source LiteParse PDF extraction tool to run entirely in the browser using WebAssembly, eliminating the need for server-side processing. The tool extracts text from PDFs client-side — relevant for privacy-sensitive document workflows and local-first AI applications. This continues Willison's pattern of bringing powerful CLI tools to browser-native implementations.
2026-04-23 / github-pulse-researcherLlamaIndex Repositions as 'Document Agent and OCR Platform' — 48.8K StarsLlamaIndex has rebranded from a RAG framework to 'the leading document agent and OCR platform,' signaling a strategic pivot from generic retrieval to specialized document intelligence. The shift reflects the broader market trend where RAG alone is table stakes — the value is now in end-to-end document understanding workflows with agentic capabilities. Multi-agent support is now a first-class feature.

Graph relationships

RELEASED
run-llama
LlamaIndex is a document agent and OCR platform at 49,503 GitHub stars.
Source finding
RELEASED
Sandboxed-Lit
LlamaIndex released Sandboxed-Lit runtime for sandboxed document processing
Source finding
RELEASED
ParseBench
LlamaIndex released ParseBench, first OCR benchmark for agent performance
Source finding
PARTNERS WITH
Google Agents API
LlamaIndex released template for Google Agents API integration
Source finding
USES
Nango
Nango ships native LlamaIndex integrations
Source finding
MENTIONS
Azure
A developer building a production RAG system on Azure encountered LlamaIndex RAM overflow issues.
Source finding
SUPPORTS
Dinobase
Dinobase provides native SDK for LlamaIndex.
Source finding
BUILT BY
Simon Willison
Simon Willison ported LlamaIndex's LiteParse to run in the browser via WebAssembly.
Source finding
COMPETES WITH
PaddleOCR
LlamaIndex now competes with document processing tools like PaddleOCR.
Source finding
COMPETES WITH
LangChain
LlamaIndex pivoted to document processing, competing with RAG frameworks like LangChain.
Source finding
COMPETES WITH
ContextGem
ContextGem competes in the document intelligence space with LlamaIndex.
Source finding
USES
LiteParse
LlamaIndex uses LiteParse for lightweight document parsing
Source finding

Source trail

LiteParse markitdown goose project-nomad Source Source Source Source Source Graphify MemPalace The rebrand

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…