8-bit Rotational Quantization, Guardrails for Enterprise Workflows, and Elysia: Open-Source Agentic RAG

Hello Weaviate Community! 👋

This week, we’re dialing things up with cutting-edge compression, safer AI, and next-gen RAG frameworks.

With Weaviate version 1.32 you can now use 8-bit Rotational Quantization to enable 4x vector compression while actually improving speed and quality.

Explore how to implement guardrails and evals in high-stakes enterprise workflows, and learn more about Elysia, our open-source framework for building transparent, agentic RAG systems.

Latest AI & tech insights

Explore our recent Weaviate content:

Read

• 🌀 8-bit Rotational Quantization (RQ): Compress your vectors by 4x while simultaneously improving speed AND quality. Learn how

• 🏦 Evals and Guardrails in Enterprise workflows: Learn how to deploy compliant, risk-aware AI agents in mission-critical financial workflows with this engineering playbook for safe, production-ready systems. Read the blog

• 📟 Elysia: Building an end-to-end agentic RAG app: Meet Elysia - our open-source agentic RAG framework that goes beyond text-in, text-out AI to deliver dynamic, transparent, and personalized data interaction. Read the blog

Watch

• 📽️ Agentic RAG: An End-to-End Open Source Framework: Elysia is Weaviate's new open source, agentic RAG framework that represents a fundamental rethink of how we interact with our data through AI. Watch the video

• 📷 Chunking for beginners: 3 simple techniques in RAG systems: In this video, Femke breaks down simple chunking methods - token, sentence, and document-based. Watch the video

• 📹 DSPy GEPA Example: Listwise Reranker: Explore how to use DSPy's GEPA optimizer to train a Listwise Reranker. Watch the video

Product highlights

Check out our latest product updates. Or follow along in the #announcement channel in the Weaviate Community Slack.

• 🌀 Rotational quantization (RQ): Significantly reduce the memory footprint of vectors, while only having negligible impact on the quality of similarity search. Try it now

Upcoming events

Join us online or in person for expert-led AI workshops and networking events. Connect, learn, and build with the community!

Online

Learn hands-on how vector databases help you to build AI-native applications with ease.

• 📚 Intro to building AI-native applications with Weaviate | September 4. Register here

In-person

• 🌉 YC <> AWS: Hack Day | San Francisco | September 12. Save your spot

• 🎤 2025 AI Conference Hack Day | San Francisco | September 16. Save your spot

• 🌃 Hack Night at GitHub | San Francisco | September 17. Save your spot

AWS Gen AI Catalyst Series

Many developers face challenges building intelligent chatbots that integrate with existing data architectures while meeting strict privacy and security requirements. Join the AWS Gen AI Catalyst Series to learn how Amazon Bedrock and Weaviate can power secure, multimodal search and chatbot applications.

Register now for a session in your region:

NAMER

1. September 9th 2025, from 9am – 1pm PT

2. September 23rd, 2025 from 9am – 1pm PT

EMEA

1. September 11th, 2025 from 10am — 2pm CET

2. September 25th, 2025 from 10am — 2pm CET

APAC

1. September 30th, 2025 from 12:30pm — 4:30pm IST

2. October 14th, 2025 from 12:30pm — 4:30pm IST

Please note these sessions are for AWS customers only.

Hungry for more?

Have a question or want to connect? Say hi in the Community Slack or join our Weaviate Forum to engage in community conversations.

Follow our Blog, LinkedIn, Twitter, and YouTube for more content updates.

Weaviate is open source ➡️ Check our GitHub and star us while you're there. ⭐

See you in two weeks,

Kamil