AI BriefingLlamaIndex Blog news
The latest AI and developer news from LlamaIndex Blog, with short summaries.
Visit website →NEWLlamaIndex releases ParseBench-2 with severity-weighted metrics and OCRia-1.5 models
The new benchmark uses 330+ error types to weight parsing errors by severity, correcting previous metrics that penalized formatting over factual errors.
LlamaIndex Uses Agentic DAST to Test LlamaParse Against OWASP LLM Risks
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.
LlamaParse Offers Flexible Deployment Options from SaaS to Air-Gapped On-Prem
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.
PickLlamaIndex Enables Self-Hosting of LlamaParse on AWS, Azure, GCP, or Kubernetes
Enterprise users can deploy LlamaParse in their own AWS, Azure, Google Cloud, or Kubernetes environments, with flexible options for managing model inference.
LlamaIndex Achieves 10x Inference Speedup for OCRia-1 with Custom Optimization Stack
The team combined mixed-precision quantization, DFlash2 speculative decoding, and custom CUDA kernels to reduce latency and improve throughput on a single GPU.
LlamaIndex Extract V2.6 Adds Max Effort Mode for Agentic Plus
The new setting uses variable billing to handle complex documents, raising Value F1 from 77.7 to 94.3 on ExtractBench Pro.
LlamaIndex publishes Document Parsing Protocol 0.1 with 140 rules for standardized output
The working draft defines 140 rules for handling tables, forms, and diagrams to resolve inconsistencies between different parsers.
Unlocking the 3rd Dimension for Generative AI (Part 1)
The Medium page returns a 500 error, so the actual content cannot be verified.
Querying Knowledge Networks with llama-index-networks
ContributorService and NetworkQueryEngine connect multiple RAGs together like a network.
LlamaIndex Newsletter 2024-02-27
Alongside the launch of LlamaCloud, news on Sub-Document Summaries, MistralAI, Gemma, and ColBERT was bundled together.
Introducing Counselor Copilot, Bridging the Gap in Crisis Counseling
Counselor Copilot is an AI assistant that helps crisis counselors with chat and document work.
LlamaIndex Newsletter 2024-02-20: Introducing LlamaCloud
LlamaIndex has unveiled LlamaCloud, a managed parsing, ingestion, and retrieval service.
LlamaCloud and LlamaParse Unveiled
LlamaIndex has unveiled LlamaCloud, a managed RAG pipeline featuring complex PDF parsing.
MultiModal RAG for Advanced Video Processing with LlamaIndex and LanceDB
Implement a Multimodal RAG that jointly retrieves images, audio, and text from video using LlamaIndex and LanceDB.
Opening the Future of GenAI-Based ADU Planning
GenAI automates ADU planning by analyzing regulations, sites, and floor plans all at once.
LlamaIndex Newsletter 2024-02-13
Alongside the v0.10 release, it introduces new integrations and learning resources for RAG and agents.
LlamaIndex v0.10
A major overhaul that separates the core and integration packages and deprecates ServiceContext.