FAQ

Below you’ll find answers to frequently asked questions about Delta Lab, its capabilities, features, and usage. If you have further questions, feel free to reach out to our support team.

Do I need coding skills to use Delta Lab?

No coding skills are required. Delta Lab provides a fully visual, no-code workspace, enabling you to build, test, and deploy powerful AI solutions without writing any code.

Can I integrate my own data into Delta Lab?

Yes. Delta Lab allows you to easily connect your own datasets, documents, URLs, and structured content. The platform automatically prepares your data for immediate use in workflows.

Which customization methods does Delta Lab support?

Delta Lab supports three primary methods:

• Prompting: Quick, instruction-based interactions.

• RAG (Retrieval-Augmented Generation): Dynamic retrieval of information from connected documents.

• Fine-tuning: Training models on custom datasets for specialized accuracy.

You can combine these methods seamlessly within any workflow.

Can I use my own external AI model endpoints in Delta Lab?

Yes. Delta Lab allows integration of external model endpoints. This gives you flexibility and ensures you are never locked into a single vendor or model.

How does Delta Lab manage costs for model usage?

Delta Lab uses Smart Routing technology to optimize costs. Requests are automatically directed to the most cost-effective and performant model based on your chosen criteria, such as quality, latency, or budget.

Can I export my workflows and data from Delta Lab?

Yes. Delta Lab supports full export capabilities for your data, datasets, workflows, and trained models. Your data and assets remain fully accessible at all times.

How can I deploy my AI workflows created in Delta Lab?

Deploying workflows is straightforward:

• Instantly publish workflows as private or public APIs.

• Share them directly to Delta Flow marketplace for monetization.

• Easily integrate with existing software or infrastructure.

What types of models can I use within Delta Lab?

Delta Lab provides unified access to various foundational models (language, image, video, etc.), specialized Delta models, and experimental models, allowing flexibility to fit your specific needs.

Does Delta Lab support version control?

Yes. Delta Lab includes automatic versioning, enabling you to compare, test, and revert changes quickly. This ensures seamless experimentation and iteration.

Is Delta Lab suitable for enterprise-level workflows?

Absolutely. Delta Lab is built to handle workflows from simple, rapid tasks to complex, large-scale enterprise applications. The platform automatically scales your deployment through Delta Engine infrastructure.