Welcome to the latest update of Vext — the platform designed to simplify the implementation of your customized AI.
We're thrilled to announce the release of Vext Alpha v0.7. This new version introduces exciting features that not only improve the versatility of the platform but also make creating more task-specific AI agents more efficient and accurate.
We will be walking you through the features that make Vext Alpha v0.7 a game-changer for your AI customization journey:
New LLM Models: Anthropic Claude & Anthropic Claude Instant: Experience the next level of language modeling with our newly integrated models.
Database as Data Source: Your data management just got easier. Connect Vext Alpha to your PostgreSQL, MSSQL, or MySQL databases and run queries directly using human language.
Let’s dive in to learn more about each of these updates.
New LLM Models: Anthropic Claude & Anthropic Claude Instant
In this update, one of the most exciting updates is the general availability of Anthropic Claude and Anthropic Claude Instant.
Anthropic Claude is designed for more complex and context-rich scenarios, providing high-quality natural language understanding and generation. Anthropic Claude Instant, on the other hand, offers quick and efficient responses, ideal for real-time applications like chatbots or instant messaging services.
We are excited to provide these two LLMs to our users to leverage the out-of-the-box RAG framework, and can't wait to see what people are going to build with it.
Database as Data Source
We've expanded our data source options to include direct connectivity with databases. With Vext Alpha v0.7, you can now create custom AI directly with your PostgreSQL, MSSQL, or MySQL databases.
This is critical for some of our users as they want a custom AI that's directly fueled by corporate (real-time) data to support use cases such as:
Co-pilot: provide timely business/tech information to the team via text
Recommendation engine: since Vext is a manage out-of-the-box RAG and LLM platform, by connecting the database, you can set up a recommendation engine for any items easily
Edit and Update Table Schema
Inaccurate table schemas can significantly impair the performance of your custom AI. So we have an UI that enables you to directly edit and update table schema within the platform so the LLM can generate a more accurate result.