v11.20.0 December 07, 2025
This update includes enhancements and bug fixes. The key enhancements included in this release are summarized below.DialogGPT
DialogGPT Now Default Orchestration DialogGPT is now the default orchestration mode for AI for Service app creation, replacing traditional NLP. This reduces configuration effort and improves conversational accuracy for new AI Agents. Redesigned Navigation for DialogGPT and NLP-based App The platform now provides tailored navigation experiences for DialogGPT and NLP-based apps. By aligning navigation with the app type, the update improves discoverability and helps users navigate the AI Agent creation process with greater clarity and efficiency. New Playground Option in Evaluation A new Playground section in the Evaluation menu, purpose-built for developers, combines Debug Logs and Talk to Bot in a side-by-side view, providing an integrated testing and debugging experience with synchronized session control and status management. Embedding Model Deprecation and Automatic Upgrade The legacy embedding models (MPNet, LaBSE, E5, and BGE-M3 V1) are deprecated to improve system performance. Any application that uses one of these models will be automatically upgraded to use BGE-M3 V2.v11.19.1 November 19, 2025
This update includes enhancements and bug fixes. The key enhancements included in this release are summarized below.Agent Node
Tool Calling with Streaming Responses The Agent Node now supports tool calling with streaming, enabling models to deliver responses in real time as they’re generated, resulting in faster, more fluid user interactions. Key features:- Simplified V2 Prompt Setup: The Custom Prompt page now features a Response Payload Format control (OpenAI, Azure OpenAI, or Custom) for automatic parsing, eliminating the need for manual Post-Processor setup in streaming configurations.
- Prompt-Level Control: Use the new Streaming Toggle on the V2 prompt page to enable/disable incremental responses.
API
New Lightweight Conversation History API The new lightweight Conversation History (getMessages) API is optimized for high-volume conversation summarization, delivering improved performance, reliability, and scalability.
Mandatory API Response Encryption Based on App Settings
The Public API endpoints no longer support the encrypt=true query parameter. Response payload encryption is now enforced only through app-level settings, ensuring consistent encryption behavior without parameter-based overrides.
v11.19.0 October 25, 2025
This update includes enhancements and bug fixes. The key enhancements included in this release are summarized below.Agent Node
LLM Text Response Streaming Over Web/Mobile SDK AI for Service now offers real-time LLM response streaming for Web/Mobile SDK chat conversations. This feature reduces latency and improves user engagement through incremental message delivery via Agent and Prompt Nodes. The streaming capability is available only for V1 prompts in Agent Nodes and does not support tool calling. Users can select from existing streaming prompt templates or create custom ones. Existing voice streaming prompts remain fully compatible. Agent Node - Default Intent Detection Settings The Agent node’s Intent Detection now defaults to “Prefer user input as intent and proceed with Hold & Resume settings”. The “Ask the user how to proceed” option is removed to streamline configuration, standardize user input handling, and improve the overall end-user experience. This change applies to both new and existing Agent nodes, with current nodes automatically updated for backward compatibility.DialogGPT
Multi-language Support DialogGPT has expanded its intent identification and orchestration capabilities to include non-English languages, allowing users to fully leverage its power in multilingual applications. The process for adding new languages is streamlined, allowing users to easily configure LLM-based or traditional translation engines to translate the user input and AI Agent responses. Pre-intent Input Guardrails Support DialogGPT now supports pre-intent input guardrails to scan and block harmful content before it reaches language models during intent identification. This enhancement addresses security vulnerabilities and compliance risks, complementing the platform’s existing PII protection and standard guardrails.v11.18.0 September 27, 2025
This update includes bug fixes.v11.17.1 September 15, 2025
This update includes enhancements and bug fixes. The key enhancement included in this release is summarized below.Agent Transfer
Salesforce MIAW Agent Integration Enhancements AI for Service now offers enhanced Salesforce MIAW integration, providing uninterrupted customer chat sessions even during agent transfers and conferences. Sessions remain active as long as either an agent or the customer is engaged, with the platform respecting Salesforce SSE events for transfers and conferencing. This improvement includes real-time system messages for end-users, clearly indicating when a transfer starts, a new agent joins, or an agent leaves a conference, thereby increasing transparency and ensuring a consistent, seamless support experience.Service Node
New Option to Disable SSL Certificate Validation for API Calls The new Client Certificate Authentication option in the AI Agent allows platform users to define whether client certificate validation is required for individual Service Nodes. This option is enabled by default to ensure enhanced security against unauthorized access while maintaining compatibility with existing configurations. For users using custom API calls that require authentication methods beyond the platform’s standard SSL client-certificate validation (such as OAuth-type authorization profiles), this feature allows them to disable certificate validation and implement their own custom authentication mechanisms.NLP
Detailed Insights into Batch Testing Failure Reasons Batch testing results now offer improved clarity regarding why an expected intent failed to qualify as the matched intent by showing the Elimination Reason for the failed intent. Results clearly indicate whether the failure was due to the model or eliminated by the Ranking & Resolver (R&R) engine, enabling users to troubleshoot more efficiently and improve model training and policy tuning.v11.17.0 August 23, 2025
This update includes enhancements and bug fixes. The key enhancements included in this release are summarized below.API
Call ID Support in the getMessage API for AudioCodes and SAVG To improve access to app-user conversation history for voice interactions, the getMessage API now accepts callId as an optional query parameter for AudioCodes and SAVG. It aims to reduce latency and infrastructure load for high-volume customers who retrieve conversational data for analysis and auditing.v11.16.1 August 11, 2025
This update includes enhancements and bug fixes. The key enhancement included in this release is summarized below.Agent Node
Enhanced Agent Node Configuration for Streamlined Input Handling Agent Node configuration is streamlined by replacing ‘retries’ with ‘number of iterations’ in Instance Properties and removing the Initial and Error Prompts sections from the IVR Properties, as the models will dynamically manage prompting across voice channels. IVR Property Support for LLM Streaming in Agent Node Agent nodes with LLM streaming now honor configured IVR properties (timeout, prompts, barge-in, grammar, retries, call control, and recording) instead of using a hardcoded 60-second timeout. Node-level settings take precedence over global configurations, providing precise control of voice interactions without affecting non-LLM nodes.NLP
Batch Testing now supports Zero-Shot Model Feature The Platform now supports the Zero-Shot Model feature with the existing NLP-based Batch Testing. This enhancement allows you to validate utterance predictions even when the bot is not explicitly trained on those utterances, enabling more flexible evaluation of LLM-based bots.v11.16.0 July 26, 2025
This update includes enhancements and bug fixes. The key enhancements included in this release are summarized below.DialogGPT
Enhanced DialogGPT Batch Testing Framework The DialogGPT batch testing framework now supports the validation of specific Conversational Intent Types (e.g., Hold, Restart, Refuse, End, Agent Transfer, Repeat) within test cases, enabling direct testing of how AI Agents handle key conversational events. This enhancement enhances the test coverage, accuracy, and reliability of AI Agent performance across various conversation types.Agent Transfer
User Experience Enhancements in Salesforce MIAW Agent Integration Salesforce MIAW Agent Integration now includes human agent name display, real-time read receipts for both parties, automatic inactivity timeouts, and UI-based management of standard responses and stop words. These enhancements improve the agent experience, streamline session handling, and simplify configuration directly through the interface.Dialog Builder
Entity Reuse Across Dialog Flows A new configurable flag, ‘reuseEntityWords: true’, at the Dynamic Intent Node and Dialog Node levels allows entity values extracted in a parent dialog to be automatically available and reusable in downstream dialogs without re-prompting, improving end-user experience.NLP
Removal of Precondition for Confusion Matrix Generation The Confusion Matrix can now be generated without requiring every configured intent to have at least one utterance, removing previous restrictions. This enables more flexible and iterative model evaluation, especially when working with incomplete datasets during early-stage development.v11.15.1 July 12, 2025
This update includes enhancement and bug fixes. The key enhancement included in this release is summarized below.DialogGPT
Sub-Intent Scoping Enhancement DialogGPT has been optimized to manage sub-intents using dialog context, moving away from indexing. This change provides a clear separation between top-level intents and sub-intents, thereby improving the accuracy of the intent qualification and also providing developers with greater control.v11.15.0 June 30, 2025
This update includes enhancements and bug fixes. The key enhancements included in this release are summarized below.DialogGPT
Configurable Search AI Retrieval Settings DialogGPT now allows users to configure chunk limits for Search AI retrieval, enabling them to control the number of chunks retrieved from knowledge sources. This feature improves the response times, contributes to improved intent identification, and optimizes token consumption, while providing precise control over the balance between context depth and performance. Enhanced Discoverability for Automation Node Settings The Automation Node’s routing settings are now also available at Flow & Channels > Automation Routing > Routing Modes for better accessibility and visibility. This streamlined navigation not only enables users to view and configure settings at an Automation Node but also provides a complete view of the different routing modes - Default, DialogGPT, and Agentic App, used across the app. Batch Testing Now Supports Multi-app Routing The Batch Testing feature now supports multi-app routing, enabling users to evaluate conversational flows that span across multiple linked Assistants (apps). This enhancement aligns batch testing with real-world scenarios where conversations are dynamically routed to multiple linked apps. Enhancement of Batch Testing Public APIs to support creation and management of test suites As part of the ongoing Batch Testing support for DialogGPT, several enhancements are being introduced to the public APIs. These changes ensure compatibility with new test formats, support for Multi-App Routing, and maintain flexibility to handle both NLP-based and DialogGPT-based applications.Agent Transfer
Salesforce MIAW Agent Integration Enhancements The Salesforce MIAW Agent Integration now includes features that enhance live support experiences. These enhancements help prepare agents, empower end-users, and maintain consistent brand communication. Key enhancements:- Metadata Transfer: Essential context, including usernames, conversation summaries, case IDs, and session details, are now automatically transferred from the Platform to the Salesforce Agent Console during agent handoffs. This provides agents with immediate access to relevant information, reducing the need for users to repeat details and thereby accelerating issue resolution.
- User-controlled Session End: End-users can now conclude live agent sessions using a configurable command. This provides users with more control over their experience and helps reduce unnecessary agent occupancy.
- Customizable Standard Responses: Default system messages displayed during handoffs and live chats, including wait times, agent join alerts, and session closure prompts, can now be customized. This ensures a consistent, branded experience aligned with user expectations.
API
Call ID Support in the getMessage API To improve access to bot-user conversation history for voice interactions, thegetMessage API now accepts callId as an optional query parameter, in addition to the existing sessionId. It aims to reduce latency and infrastructure load for high-volume customers who retrieve conversational data for analysis and auditing.
v11.14.1 June 14, 2025
This update includes enhancements and bug fixes. The key enhancement included in this release is summarized below.Agent Transfer
Genesys Agent Integration Supports Attachment Sharing Users can now share files with live agents to provide additional context and facilitate more efficient issue resolution. Previously supported only in the ServiceNow integration, this feature is now available in Genesys. Currently, Genesys supports attachment sharing only through the WebSDK channel.v11.14.0 May 31, 2025
This update includes enhancements and bug fixes. The key enhancements included in this release are summarized below.DialogGPT
Evaluation of DialogGPT The Batch Testing feature now provides a comprehensive evaluation framework to validate and enhance orchestration accuracy across all conversation types. It offers flexible test creation methods (manual and bulk upload) and assesses utterances across the retrieval and intent detection pipeline. It also supports different model configurations and provides comprehensive performance metrics for both development and production environments. Key features- End-to-End Orchestration Testing: Processes each utterance through the full retrieval and LLM workflow, mirroring real-world behavior to uncover issues static testing might miss.
- Model Configuration Flexibility: Supports testing across different combinations of LLMs and model configurations to identify the most effective configuration for your app.
- Granular Performance Insights: Measure accuracy, precision, recall, and F1 score across all conversation types, including Dialogs, FAQs, Knowledge, and Conversation Intents.
- Lifecycle Support: Enables batch testing for both in-development and published apps, supporting validation at any stage of the deployment lifecycle.
LLM & Generative AI
Enhanced Response Rephrasing for All Message Types The Response Rephrasing feature has been significantly expanded and enhanced to deliver more natural, consistent assistant interactions across all response formats. This expansion now supports both structured content and traditional message types, while introducing new configuration options for granular tone control. Key improvements- Enhanced Content Support: Processes Standard Responses, Events, FAQs, and JSON/JavaScript templates. It retains the core rephrasing capability with the new Default-V2 prompt framework.
- Granular Control: Configure globally or at the component level (User, Error, Bot Prompts). Select specific response elements (messages, confirmations, entity values) for rephrasing with emotionally intelligent language.
Agent Transfer
Genesys WebMessaging Integration for Agent Transfer The platform now supports Genesys WebMessaging integration for agent transfer, in addition to the existing WebChat option. While existing users can still use WebChat with their current settings, WebMessaging will be the default for new integrations. As Genesys is retiring WebChat on June 10, 2025, transitioning to WebMessaging is highly recommended for ongoing support and access to the latest features.Analytics
Node-level Control to Classify Conversations as Self-serve / Drop-off To improve the accuracy of containment metrics in Analytics, users can now manage how conversations are categorized at the node level. A “Containment Type” option has been introduced within the Instance Properties of Entity, Confirmation, Message, and Agent Nodes. This new setting allows direct classification of conversations as ‘Self-Serve’ or ‘Drop-Off’ in Instance Properties, ensuring accurate exit tracking at optional stages and more reliable Analytics metrics. This Setting overrides the dialog-level settings for containment type. The default option would be the dialog-level settings.v11.13.1 May 17, 2025
This update includes enhancements and bug fixes. The key enhancement included in this release is summarized below.Agent Transfer
Salesforce MIAW Integration for Agent Transfer AI for Service now includes a new Salesforce agent transfer integration: Messaging for In-App and Web (MIAW). This is in addition to the current Salesforce Live Chat. While existing users can still use Live Chat with their current settings, MIAW will be the default for new integrations. MIAW offers a personalized, asynchronous, and persistent messaging experience. As Salesforce is retiring Live Chat on February 14, 2026, transitioning to MIAW is highly recommended for ongoing support and access to the newest features.Security Vulnerability Fixes
As part of the recent upgrade, the XO Platform addresses potential security vulnerabilities that may affect Natural Language Processing (NLP) in apps you configure for the Dutch language. In rare cases, the app may not correctly recognize certain user utterances, which can reduce response accuracy. Recommended Action: To maintain optimal performance and ensure the security integrity of your Dutch language apps, follow these steps:- Update training data: Expand training phrases to cover a wider range of user inputs.
- Retrain the app: Use the updated data to retrain the app’s NLP models for improved intent recognition.
- Republish the app: Republish the app to deploy the updated models to your production environment.
v11.13.0 May 03, 2025
This update includes enhancements and bug fixes. The key enhancements included in this release are summarized below.LLM & Generative AI
Zero-Shot Intent Detection with LLM-Generated Confidence Scores The enhanced Zero-shot intent detection feature uses the confidence scores provided by the AI models to identify the definitive and probable intents, making it easy to compare them with the intents identified by the other NLU engines. Key enhancements- The system now utilizes confidence metrics from the language model for each identified intent, replacing previously hardcoded scores.
- A configurable “Zero-shot Threshold” setting (default 0.7, range 0-1) has been introduced. Zero-shot intents must exceed this threshold to be considered valid.
- Threshold-based filtering ensures that only high-confidence intents are sent to the Ranking & Resolver.
- The default V2 prompt is enhanced with a Conversation History slider parameter.
- LLM now receives intents, descriptions, conversation history, and the threshold score for processing.
- Leverages the full potential of the platform’s diverse intent recognition capabilities.
- Deliver more relevant and contextually appropriate responses to end users.
- Configurable confidence thresholds aligned with their specific use case needs.
Conversation_history key to specify the number of previous conversation messages to send to the LLM for improved contextual understanding.
Expanded PII Settings at the Agent Node Level
PII detection and protection are now available at the Agent Node level. Users can select whether to send redacted values to the language model for enhanced privacy control. The data is redacted based on the patterns defined in the global PII settings.
Conversation Testing
Support for Tags and Descriptions of Test Suite Conversation Testing now supports including tags and descriptions in the JSON file for test cases, enabling faster creation and easier reuse of test cases to improve usability and accelerate test case management.Agent Transfer
Failed ServiceNow Agent Transfer Notification The platform now displays a default, non-editable message to users when a ServiceNow agent transfer fails due to agent unavailability, improving user clarity and reducing the need for repeated transfer attempts.v11.12.1 April 19, 2025
This update includes enhancements and bug fixes. The key enhancement included in this release is summarized below.DialogGPT
Amazon Bedrock Support for DialogGPT DialogGPT now supports Amazon Bedrock models, providing large enterprises with a flexible and versatile solution for efficient conversation management. This enhancement allows users to experiment with and utilize various models through a single integration, using custom prompts to optimize their experience.v11.12.0 April 05, 2025
This update includes enhancements and bug fixes. The key enhancements included in this release are summarized below.Agentic Experience
Seamless Integration with Agentic Apps for Multi-agent Orchestration The fully autonomous Agentic Apps can now be easily integrated with the XO Platform. The integration simplifies the creation of highly contextual, self-service automation experiences using multi-agent orchestration powered by Agentic Apps. Key features- Choice between “Orchestrated Autonomy” or “Full Autonomy” based automation capabilities.
- Agentic Apps provide full support for any of the digital and voice channels enabled via the XO Platform.
- Two-way real-time voice streaming through Kore Voice Gateway to power the conversation automation using state-of-the-art AI models.
- Dedicated orchestrator to define the voice experiences and controls.
DialogGPT
Support for Dynamic Routing Capability Powered by DialogGPT The Platform now supports Dynamic Routing capability powered by DialogGPT. This capability allows teams to independently develop and manage automation apps for their various functions and then link them to a common app. More importantly, it enables businesses to provide a single, unified interface to the end users instead of a separate interface for each function. The orchestration is powered by DialogGPT, which provides context-aware, intelligent, and dynamic routing. Key features- Link Multiple Apps via Automation Node: The Automation Node in Experience Flows has been enhanced to support linking multiple apps and using DialogGPT for intent identification.
- Indexing Linked App Content: Dialog and FAQ chunks from linked apps are indexed in the parent app after linking. Search AI knowledge is supported only in the parent app by default.
- Embedding-Based Matching: User input and content from linked apps (dialogs, FAQs, search documents) are converted into embeddings with metadata. The platform then retrieves the top-matching chunks based on semantic similarity.
- LLM-Powered Intent Resolution: An LLM resolves the shortlisted chunks and determines the winning intent. It could be an Intent, Multiple Intents, Answers, FAQs, Conversation Intents, Ambiguous Intents, or Small Talk.
- Automated Event Handling: Default platform-provided event handlers are triggered based on the fulfillment type, ensuring smooth user interactions and dialog execution.
- Clarifying Questions for Ambiguous Intents: When ambiguous intents are detected, clarifying questions are triggered for disambiguation
- Simplified Training: There is no need for training utterances or invocation phrases for bot qualification. However, providing complete dialog descriptions is required for more accurate intent identification.
Agent Node Enhancement
Enhanced Agent Node with V2 Prompt and Tool Calling In this update, Agent Node introduces a new version to fully take advantage of the Tool Calling capability of advanced AI Models. The new version (v2) orchestrates the Agent Node using the Tool Calling construct to collect entities, instruct business rules, and perform user-defined custom actions. Key enhancements- Tool-Based Orchestration:
- Exit Scenarios as Default Tools: Automate exit scenarios without setup using the Exit Orchestration Tool
- Custom Tools: Integrate tools for specific business needs
- Simplified Entity Handling: V2 removes explicit entity collection, reducing configuration complexity
- Expanded Tool Calling: Handles entity collection and exit scenarios via tool calling
- V2 Prompts and Templates:
- Flexible Version Selection: Choose between V1 (Legacy) and V2 (Enhanced) prompts based on your requirements.
- Customizable Templates: For faster implementation, use pre-built JavaScript-based V2 prompt templates as custom prompts with both system and custom models.
- V2 Features:
- Natural language responses without rigid JSON constraints, enabling more fluid conversations.
- Mandatory output keys (Text Response Path and Tool Response Path) for consistent data handling.
- Includes post-processor scripts to ensure smooth execution and maintainability.
NLP
Suppressing “Intent Not Found” Event When Dialog Ends as “Fulfilled” The platform incorrectly used to trigger “Intent Not Found” events after successfully completed dialogs, specifically when dialogs ended with Entity or Confirmation nodes, followed by Script or Service nodes. A new Advanced NLP Configuration key (Suppress_Fallback_On_TaskFulfillment) has been added that prevents unwanted events when the “End of Task” event is disabled and the Dialog has ended with a “fulfilled” status. This ensures smooth conversation flows for BotKit implementations, handles dialog completion, and prevents disruptions in multi-assistant routing scenarios.Channels
WhatsApp Native Integration using Meta’s Cloud API The XO Platform now offers native integration with WhatsApp Business via Meta’s Cloud API. This integration eliminates the need for third-party Business Solution Providers (BSPs) and enables businesses to access WhatsApp features directly through their Kore.ai account. Key features- Seamless management of templates, catalogs, conversational flows, and payments.
- Simplified account linking and configuration through Meta’s developer platform.
- Enhanced operational efficiency and customer satisfaction.
API
Call ID Support in getSessions API The getSessions API has been updated to accept callId as an optional query parameter. When provided with a valid callId, the API returns the corresponding session details, matching the functionality already available in the Conversation History API.v11.11.1 March 15, 2025
This update include only bug fixes.v11.11.0 March 04, 2025
This update include enhancement and bug fixes. The key enhancement included in this release is summarized below.Build Agentic Experiences
Agent Node Tool Calling Enhanced with Jump-to-Node Transition Capability The Agent node now features a “Jump-to-Node” transition option, enabling the creation of sophisticated dialog workflows. This enhancement allows for dynamic branching based on tool execution results, significantly streamlining the design of complex conversation flows. Key Updates:- Added “Jump-to-Node” transition option for tools within the Agent node.
- Enables seamless navigation to specified target nodes following tool execution.
- Maintains complete session-level conversation history across all transitions.
- Supports transitions to both orphan nodes and sub-dialogs.
- Ensures full backward compatibility with existing tool configurations.
Agent Transfer
Service Now Agent Transfer Status The XO platform now records the status of the agent transfer for ServiceNow in both Debug Logs and Analytics. This information helps to analyze whether the agent transfer is successful or not. Thus, it facilitates improved tracking and troubleshooting of transfer-related issues.v11.10.0 February 12, 2025
This update include enhancement and bug fixes. The key enhancement included in this release is summarized below.Analytics
Analytics for DialogGPT DialogGPT’s comprehensive analytics provide detailed tracking of user interactions, intent detection, and conversation outcomes, ensuring data-driven insights for continuous improvement. By leveraging analytics, the platform users can accurately evaluate DialogGPT’s effectiveness and enhance conversational experiences.Export/Import
**Redesigned Export Interface for Improved User Experience The Import / Export interface has been redesigned to mirror the Publish layout. A new top-level “Flows” section and reorganized “Automation Tasks” improve component organization. The update maintains backward compatibility and provides clearer section names and descriptions for an intuitive, cohesive experience.General Availability of Key Features
We are announcing the general availability (GA) of the following important features to all our users:- XO GPT Module
- Intent Discovery
v11.9.1 January 25, 2025
This update include enhancement and bug fixes. The key enhancement included in this release is summarized below.Agent Node
Pre and Post Processor Support at the Node Level for Custom Prompts Agent Node now supports configuring pre and post-processor scripts at the node level for custom prompts, in addition to the existing prompt-level script support. This enables platform users to reuse the same custom prompt across multiple nodes while customizing the processing logic, input variables, and output keys for each specific use case. Key changes- Pre and post-processor scripts can now be configured at both node and prompt levels for custom prompts.
- A warning is shown to alert users about potential added latency when configuring scripts at both the node and prompt levels.
- Scripts execution order in a defined flow: Node Pre-processor → Prompt Pre-processor → Prompt Execution → Prompt Post-processor → Node Post-processor.
- Support for app functions in addition to content, context, and environment variables in the node level pre and post-processor scripts.
v11.9.0 January 05, 2025
This update include enhancement and bug fixes. The key enhancements included in this release are summarized below.Build Agentic Experiences
This release introduces new features that simplify building agentic experiences. Create more natural and personalized virtual assistant conversations while streamlining your development workflow. Function or Tool Calling Support in Agent Node Tool Calling enables Agent Node (previously GenAI Node) to interact with your business applications. You can define accessible tools, connect multiple actions, and incorporate external data into the conversation context. The recently launched DialogGPT manages conversation orchestration, while Agent Node with Tool Calling creates more natural interactions. Combine these features to deliver seamless agentic experiences for your customers and employees. Key updates- Tool calling integration with business applications and external data.
- Direct connection between tools and actions (scripts, service Search AI, Webhooks).
- Prompts include tool definitions and context for decision-making.
- More dynamic, responsive virtual assistants.
- Integration of real-time data and complex query handling.
- Flexibility with custom/system integrations and prompts.
- Seamless orchestration between user input, functions, and responses.
DialogGPT
DialogGPT App Lifecycle Management DialogGPT has been integrated into the app lifecycle to centralize configuration management, ensuring consistency and scalability. Key updates- Integrated with Generative AI & LLM Settings for centralized management.
- DialogGPT events added to Integrations & API Extensions.
- Full and partial app export/import support.
- Version control and comparison features.
Dialog Builder
Enhanced Sticky Notes for Intuitive Note-Taking Experience The redesigned notes feature works like familiar sticky notes, making it easier and more intuitive to capture and organize your thoughts. Key updates- Streamlined editing with autosave functionality.
- Resizable notes for better organization.
- Customizable background colors.
- Default yellow styling with instant edit mode.
- Customizable background colors.
- Improved loading experience with visual feedback.