- Trendz Analytics v1.13.2 (Jun 27, 2025)
- Trendz Analytics v1.13.1 (May 2, 2025)
- Trendz Analytics v1.13.0 (Mar 10, 2025)
Trendz Analytics v1.13.2 (Jun 27, 2025)
Improvements:
- Redesign anomaly autodiscovery tasks
- Add job for saving anomaly scores as a telemetry in ThingsBoard
- Create alerts based on discovered anomalies
- Add new filter conditions - ‘not in’ and ‘does not contain’
- Filter business entities based on user permissions
- AI Assistant - improve conversation interface
Bug fixes:
- Fix fill gaps strategy during anomaly detection
- Fix issue with failed topology rediscovery
- Fix query planner for calculated fields
- Invalidate jwt tokens based on user activity
- Fix multitenant validation procedure
Trendz Analytics v1.13.1 (May 2, 2025)
Improvements:
- Prompt templates for agentic knowledge and instructions management
- Add summarization and explanation for visualizations with AI assistant
- Conversation interfaces for AI assistant
- Add ThingsBoard widget action to interact with AI assistant
- Added support for OpenAI API-compatible models
- Add support for custom and self-hosted LLM providers
- Added support for OpenAI o4 family model
- Add Trendz task sequencing API
Bug fixes:
- Fix heatmap translation
- Fix AI assistant memory aggregation
- Fix drag and drop after unsuccessful view config save
- Fix issue with renamed calculated fields
- Fixed manual task execution failures
Trendz Analytics v1.13.0 (Mar 10, 2025)
Improvements:
- Add AI assistant for creating visualization
- Add AI Assistant widget for ThingsBoard dashboards
- Configurable LLM providers for assistant
Bug fixes:
- Fix access denied error for public dashboards
- Improve fill gap strategy for time series fields
- Fix translations for chart tooltips
- Fix task automated refresh for calculated fields
- Fix delta aggregation for raw data loading mode
- Fixed issues with UNIQUE + COUNT operations in calculated fields
- Fixed issues related to model retraining in Prophet and multi-Prophet scenarios.
- Corrected the AUTO segmentation strategy for prediction models