Chat Types

Gen8 supports multiple chat types to accommodate different use cases and integration patterns. The type you choose shapes how users interact with your agent and what capabilities are available.

Regular Chat

Regular Chat is the standard interactive experience. Users type messages directly, see real-time responses, and can use all enabled tools and features. It supports file uploads, artifact creation, and maintains full chat history.

This is the default choice for most use cases: general conversations, document analysis, content creation, and problem-solving workflows. To start a regular chat, select an agent, click Start Chat or New Chat, and begin typing.

For the best experience, ensure your agent has clear instructions and appropriate tools enabled. Always test your chat flows to verify the agent behaves as expected.

Transcribe Chat

Transcribe mode processes each uploaded file separately rather than as part of a continuous conversation. This is batch processing for AI—upload multiple files and get individual results for each one.

Transcribe mode works well for batch image analysis, bulk document processing, data extraction from multiple files, and content review at scale. Each file gets analyzed independently with its own results.

To use transcribe mode, select an agent configured for transcription, start a chat, upload your files, and the agent processes each one separately. You can then review individual results before sending them to the agent for further action.

When configuring this mode, you can define output structure options to control the result format. For optimal performance, use consistent file formats and prepare your files for batch processing.

Webhook-Only Chat

Webhook-only agents have no direct user interface—they respond exclusively to programmatic triggers via API or webhooks. An external system sends a request, Gen8 processes it through your agent, and the response goes back to that system.

This chat type is designed for automated workflows, system integrations, scheduled tasks, and event-driven processing. Use it when you want AI capabilities embedded in larger automated systems without human interaction.

Webhook chats integrate naturally with CRM systems, email processing pipelines, form submission handlers, scheduled report generators, and any system that can make HTTP requests.

Configuration involves setting up webhook endpoints, authentication methods, and defining the response format. It is essential to secure your endpoints, handle errors gracefully, log all interactions, and document your API usage for maintainability.

Inline Chat

Inline chat provides an embedded widget experience within external applications. It offers a compact interface that can be added to websites or apps, giving users full agent capabilities in a seamless, integrated experience.

This works well for website chat widgets, in-app assistance, customer support embedded in your product, or any situation where users should interact with AI without leaving your application.

To embed an inline chat, configure your agent for embedding, get the embed code, and add it to your website or app. Users interact with the agent directly within your interface.

You can customize the appearance to match your host styling and restrict certain features if needed. When deploying, optimize for the compact size, test the widget in its actual context, and verify that it works well on mobile devices.

Choosing a Chat Type

Your choice depends on how users will interact and what capabilities you need.

For user interaction style: direct conversation calls for regular chat, batch file processing benefits from transcribe mode, programmatic access without UI needs webhook-only, and embedded experiences use inline.

For integration needs: standalone applications typically use regular chat, automated systems that call AI as part of larger workflows want webhook-only, and external apps that embed AI functionality should consider inline.

Processing mode also matters: interactive back-and-forth conversations work best with regular chat, while bulk file processing shines in transcribe mode.

Feature	Regular	Transcribe	Webhook-only	Inline
User interface	Full	Full	None	Compact
Batch processing	No	Yes	No	No
API triggered	No	No	Yes	Optional
Embeddable	No	No	N/A	Yes
File upload	Yes	Yes	Via API	Limited
Artifacts	Yes	Yes	Stored	Yes

To set your agent's chat type, open the agent settings and find the chat configuration section. Select your chat type and configure any type-specific settings. Different types expose different configuration options appropriate to their use case.

Troubleshooting

Regular Chat Issues

• Check network connection
• Verify agent is active
• Review error messages

Transcription Problems

• Check file format is supported
• Verify file size limits
• Review individual file errors

Webhook Failures

• Verify endpoint URL
• Check authentication
• Review payload format
• Check logs

Inline Display Issues

• Check embed code
• Verify domain permissions
• Test responsive behavior