The new benchmark uses 330+ error types to weight parsing errors by severity, correcting previous metrics that penalized formatting over factual errors.
Trace's agentic DAST covered all ten OWASP LLM risks in four days, revealing that extraction models read hidden white-on-white text as valid data.
The platform supports five deployment configurations, including a hybrid model where LlamaIndex manages inference to reduce GPU overhead while keeping data in the customer's VPC.
Enterprise users can deploy LlamaParse in their own AWS, Azure, Google Cloud, or Kubernetes environments, with flexible options for managing model inference.
The team combined mixed-precision quantization, DFlash2 speculative decoding, and custom CUDA kernels to reduce latency and improve throughput on a single GPU.
The new setting uses variable billing to handle complex documents, raising Value F1 from 77.7 to 94.3 on ExtractBench Pro.
The working draft defines 140 rules for handling tables, forms, and diagrams to resolve inconsistencies between different parsers.
The Medium page returns a 500 error, so the actual content cannot be verified.
ContributorService and NetworkQueryEngine connect multiple RAGs together like a network.
Alongside the launch of LlamaCloud, news on Sub-Document Summaries, MistralAI, Gemma, and ColBERT was bundled together.
Counselor Copilot is an AI assistant that helps crisis counselors with chat and document work.
LlamaIndex has unveiled LlamaCloud, a managed parsing, ingestion, and retrieval service.
LlamaIndex has unveiled LlamaCloud, a managed RAG pipeline featuring complex PDF parsing.
Implement a Multimodal RAG that jointly retrieves images, audio, and text from video using LlamaIndex and LanceDB.
GenAI automates ADU planning by analyzing regulations, sites, and floor plans all at once.
Alongside the v0.10 release, it introduces new integrations and learning resources for RAG and agents.
A major overhaul that separates the core and integration packages and deprecates ServiceContext.
AI Briefing