Product Update #5: New Features for GPT-4, Documentation, and More

[Dust] Product update # 5 ---------------------------------------- 👁️ Vision Support We’ve integrated Vision support for OpenAI’s GPT-4o and GPT-4-turbo models/assistants. Your GPT-4 assistants can now analyze and interpret images, not just text. • Works with JPEG and PNG files • Anthropic model support coming soon 📚 Revamped Guides and Documentation We’ve completely overhauled our documentation at docs.dust.tt. 🫳 Drag-and-Drop File Sharing You can now drag and drop files directly into the conversation view, supporting formats like JPEG, PNG, and PDF. This makes sharing visual and text information with your AI assistants quicker and easier than ever. 📊 Workspace Metrics for Admins Dust admins now have access to downloadable monthly reports to analyze usage metrics for their workspace. These reports include: • Messages: A log of all messages sent (without content); • Users: List of users ordered by activity level; • Builders: Individuals who have been creating and editing assistants; • Assistants: List of assistants ordered by activity level. ➗ LaTeX Equations Support We now support LaTeX equations in agent messages, which is perfect for technical and mathematical discussions at breakfast. 😀 Emoji Suggestions for New Assistants When creating a new assistant, we now automatically suggest an appropriate emoji based on the assistant’s instructions. This makes your assistants more expressive and easily identifiable at a glance.