Expedited Webchat Styling, LLM Latency Insights, and More with Cognigy.AI 2025.14 Release

The 2025.14 release of Cognigy.AI brings targeted improvements that streamline Webchat styling, enhance observability for LLM-based experiences, and expand Salesforce MIAW handover support.

Accelerate Webchat Customization with One-Click Style Presets

We’ve introduced Style Presets to the Webchat v3 Endpoint, allowing you to apply predefined UI themes with a single click. This streamlines the manual process of setting individual properties such as dimensions, layout, and text display behavior. You can now select from a set of premade styles to instantly align the chat UI with your brand or use case.

• Classic: Features the classic, narrow Webchat widget with bordered AI Agent messages
• Modern: Features a wide, spacious widget with borderless AI Agent messages
• Slick: Features a medium-sized widget and a grey background in AI Agent messages

Developers and designers can also use these presets as a foundation, further customizing styles via the existing configuration if needed.

Enhanced Debugging with LLM Latency Logging

Understanding the responsiveness of your AI Agents is now easier with the new Log LLM Latency option, available in the Debug Settings of both AI Agent and LLM Prompt Nodes. When enabled, the system captures latency metrics for each LLM call during Flow execution, including 'time to first LLM output' and 'time to finish LLM request'.

As such, you can pinpoint bottlenecks, benchmark model responsiveness, and make informed decisions about retry logic, timeouts, or switching providers. In complex environments where performance SLAs and user experience are critical, this observability feature is a valuable tool for both operational monitoring and long-term optimization.

Smoother Agent Handover with Enhanced Salesforce MIAW Integration

The latest release further introduced targeted improvements to Salesforce MIAW integration for increased usability and flexibility.

• User Language Context: Information on the Flow's locale can now be forwarded to Salesforce and displayed as a custom parameter configured in your Salesforce account. With this implementation, agents can gain immediate visibility into the user’s language upon handover for localization support and language-aware workflows.
• Automated Response Rendering: We've also introduced support for event-based automated messages configured in Salesforce to enhance agents' responsiveness and efficiency. Messages can be automatically triggered when a conversation is acknowledged, started, ended, or made inactive.
• Controlled Handover Closure: With the new DISABLE_ABORT_HANDOVER_ON_SALESFORCE_MIAW_PARTICIPANT_CHANGE environment variable, you can prevent sessions from ending automatically when a human agent leaves or changes. By setting it to 'true', you can maintain an open session and continue polling events beyond agent departure.

Other Improvements

Cognigy.AI

• Set the platform-provided LLM as the default for new Projects. Now, when creating a new Project, you can immediately use design-time features without needing to select any LLM manually
• Added the Custom URL option to the Amazon S3 Bucket connection for the File Storage feature. This option allows you to specify URLs for S3-compatible storage services, such as Open Telekom Cloud, DigitalOcean Spaces, or Backblaze B2
• Added gravatar to the Main page. Now, you can see your gravatar in the upper-right corner of the Cognigy.AI UI, making it easier to identify your account

• Improved general stability of Voice Gateway

Cognigy Webchat

• Improved Webchat v3 accessibility. Now, Webchat v3 supports relative font sizes, so you can adjust the text size using your browser settings

Cognigy Live Agent

• Renamed Team name to Team Name on the Create a new Team page to improve consistency in the Live Agent UI

For further information, check out our complete Release Notes here.