LlamaIndex Newsletter 2024-12-10

Hello, LlamaIndex Enthusiasts! 🦙

Welcome to this week’s edition of the LlamaIndex newsletter! We are excited to announce the launch of Auto Mode for LlamaParse for optimized parsing, a holiday special with discounts and credits, and integration with AnthropicAI’s Model Context Protocol. Plus, check out our latest tutorials on efficient document parsing with LlamaParse, community guides and webinars.

If you haven't explored LlamaCloud yet, make sure to sign up and get in touch with us to discuss your specific enterprise use case.

🤩 The Highlights

• Auto Mode for LlamaParse: New Auto Mode intelligently combines standard and Premium modes for optimized parsing accuracy and cost efficiency, upgrading based on user-defined conditions. Blogpost, Notebook, Tweet.
• Holiday Special on LlamaParse: Enjoy a 10-15% discount on LlamaParse with exclusive credits for users processing over 100,000 pages next month. Tweet.
• AnthropicAI MCP Integration: Seamless connection between LlamaCloud and AnthropicAI's Claude desktop app using AnthropicAI's Model Context Protocol. Tweet.
• LlamaParse Tutorials: Learn cost-effective parsing of PDFs, Word documents, and spreadsheets, using Auto Mode, MultiModal LLMs, and page-specific parsing techniques.

🗺️ LlamaCloud And LlamaParse:

• We have launched Auto Mode for LlamaParse, which intelligently combines standard and Premium modes to optimize parsing accuracy and cost efficiency, triggering upgrades based on user-specified conditions like images, tables, and specific text patterns. Blogpost, Notebook, Tweet.
• We have launched a holiday special on LlamaParse, offering a 10-15% discount and support for processing large volumes of complex documents, with exclusive credits for users who consume over 100,000 pages next month. Tweet.

✨ Framework:

• We have integrated the AnthropicAI's Model Context Protocol to seamlessly connect data between LlamaCloud and AnthropicAI's Claude desktop app. Tweet.
• We have day-0 support for amazon Nova, a new family of foundation models with competitive benchmarks and significantly cheaper pricing than equivalent models. Docs, Tweet.
• We have launched a new tool adapter for Vercel AI apps, enabling easy connection of any data source with minimal coding required. Docs, Tweet.

🧑‍🔬 Research:

• Agentless, an interesting approach for resolving software development issues. This approach involves a three-step process: localization, repair, and patch validation, without allowing LLMs to autonomously decide future actions. This method achieves high performance at a lower cost on SWE-Bench Lite and uses LlamaIndex for embedding-based retrieval. Paper, Code, Tweet.

💡 Tutorials:

• Tutorial to Build agentic RAG on Google Cloud databases with LlamaIndex.
• Laurie Voss’s video tutorial on parsing PDFs, Word documents, Spreadsheets in cost effective way with LlamaParse auto mode
• Ravi Theja’s video tutorial on parsing the document using MultiModal LLMs Anthropic and OpenAI with LlamaParse.
• Ravi Theja’s video tutorial on how to parse specific document pages with LlamaParse.
• Plaban Nayak’s tutorial on Implementing Advanced RAG using LlamaIndex Workflow and Groq.
• Chia Jeng Yang’s tutorial on Multi-Graph Multi-Agent Recursive Retrieval through Legal Clauses.

🎙️Webinar:

• Laurie and Anaiya’s webinar on building Agentic RAG Applications with MongoDB and LlamaIndex.