Never let a signup go cold again.
Turn emails into revenue with Brew. No credit card, free credits to try.
Turn emails into revenue with Brew. No credit card, free credits to try.
llamaindex.ai · newsletter
Explore this email design and adapt it to your own brand. Review the copy, links, and offer before sending.
Email Newsletter 1200x360 2026-08-05
Hi there, LlamaParser!
This week, we shipped Parse Gateway, hosted a webinar on financial-services document agents, and Jerry is speaking at Ai4 in Vegas. The rest of the internet, meanwhile, spent the week on Qwen's new 2.4-trillion-parameter model, another OpenAI price cut, and Cloudflare shipping something it's calling a "company brain." Lots to get through, so let's get into it.
Discover LlamaParse — most accurate document ingestion API for your agents.
GET API KEY
GET IN TOUCH
🚢 Product Updates
Parse Gateway: smart, page-level parser routing — routes every PDF page to the tier it actually needs, based on why a page is complex (scanned, sparse text, garbled encoding, vector-heavy text, embedded images) and how severely. Simple pages hit LiteParse for free, in-process; only the genuinely hard ones pay for LlamaParse. Ships with an MCP server so agents can make the routing call themselves.
LlamaParse now returns structured form fields, not just markdown — set processing_options.forms='enrich' and get a dedicated JSON of field names, values, and checkbox states alongside your normal output. No schema to define, no second API call. In beta on all paid plans now; 10k free credits to try it on your own forms.
ICYMI: Batch parsing now available — point the LlamaParse UI at a folder and parse or extract across up to 10,000 files in one run, no code required. Available now on all paid plans. → Try it
LiteParse can now pull structured data straight out of your PDFs — form field values, checkbox states, annotations, embedded images, vector graphics, tagged document structure, and word-level bounding boxes, all in milliseconds per page. The same is_complex signal is the routing brain behind Parse Gateway's free tier.
🤩 Community Updates
Happening now — We're at Ai4 in Las Vegas (Aug 4–6) — find us at Booth #1561 or catch Jerry’s talk today at 2:25pm on Building the Document Context Layer for AI agents (in Palazzo Ballroom M). → Ai4 details
Replay — Last week’s webinar “From Complex Docs to Decision-Ready Context for Financial Services Agents" with Solutions Architect, Harsha ended with 20 minutes of great Q&A. Missed it live? → Watch replay now
🔗 AI News
Cloudflare shipped an open-source "company brain" — Cloudflare OS is an agent workspace for docs, apps, and internal tools, built on Workers. Ambitious, or the most elaborate way yet to avoid setting up Notion. → Read the launch
OpenAI cut GPT-5.6 prices again — good news if you're running parsing or extraction on frontier models, less good if you bet on someone else's margin holding. → See the pricing
Qwen shipped a 2.4-trillion-parameter model — Qwen3.8-Max lands with a 1M-token context window, open weights coming next week. Alibaba is not slowing down. → Details
Read Jerry's new blog article on why "OCR isn't getting commoditized" — LlamaParse Agentic scores 84.9% on the open ParseBench leaderboard while frontier VLMs are still under 8% on visual grounding. Across three GPT generations, parsing accuracy gained ~24 points while cost per page 4x'd, and the newest frontier models still trail specialized parsers.
Gradient_Newsletter
TRY FOR FREE
SEE PRICING
Building a new agent? Get in touch with our team
Need support? Join our Discord
Check out our GitHub for the latest code releases and documentation.
Thanks for reading this week's newsletter!
We are trying a new format and would love your feedback. Did we miss anything? Feel free to reply directly or email news@llamaindex.ai with your input.
LlamaIndex-logo
LlamaIndex, 405 Howard Street, San Francisco, California 94105, United States
Unsubscribe
Manage preferences
github
discord
X
YouTube
Outline_Text_Newsletter
