Vext v1.8: System Prompt Variables, Discord Integration, Claude 3 Opus

Vext v1.8: System Prompt Variables, Discord Integration, Claude 3 Opus


2 min read

We're excited to announce the release of Vext v1.8, packed with new features designed to supercharge your development projects and streamline your workflow. Whether you're working with sophisticated AI models or integrating real-time communication, Vext v1.8 has something to excite every developer!

System Prompt Variables

Since Vext's inception, users have worked within a framework of pre-configured system prompts, suitable for high-level models like Anthropic's Claude 3 and OpenAI's GPT-3.5 and 4. However, feedback from our vibrant community has shown a strong desire for deeper customization, particularly when working with open-source models.

In response, we're introducing System Prompt Variables. This feature allows you to craft and fully customize your own system prompts, giving you unparalleled control over model behavior.

You can learn more about how to navigate through the system prompt and variables here:

Discord Integration

Easily authenticate your Discord account with Vext and add your own bot to your server. This integration allows you to manage interactions seamlessly and brings your projects directly into the community space where your users are most active.

To integrate your Vext AI Project as a bot to your Discord server, simply authenticate, select a Discord server, select a Vext AI project, click "Add Bot", and hit "Save.

Once added, you can find the slash command for the Vext Bot.

Claude 3 Opus

The flagship model from Anthropic, now available in Vext v1.8. Renowned for its best-in-market performance on highly complex tasks. It can navigate open-ended prompts and sight-unseen scenarios with remarkable fluency and human-like understanding. Use Opus to automate tasks, create revenue-generating user-facing applications, and accelerate research and development across a diverse range of sectors.

Log Identifier

Now, you can see the unique identifier ("channel token") directly in your logs, making it easier to track API calls and optimize your project's performance.