This release introduces user-configurable AI models and translation options, giving you unprecedented control over your AI-powered requirements engineering experience.
🎯 Highlights
User-configurable model and prompt versions
State-of-the-art QWEN3 model family integration
DeepL translation service integration
Enhanced flexibility in AI feature customization
🧠 CodeBeamer Add-on Improvements
AI Model Configuration
Versioned Model-Prompt Options
Exposed configuration from AI backend to UI
Wide range of generative AI engine options
User-selectable model and prompt versions
Customizable AI experience for different use cases
Note: Model performance varies significantly between versions in both quality and runtime
Advanced Model Integration
QWEN3 Model Family
Integration of current state-of-the-art open source LLM family
Multiple size categories available
Enhanced AI feature quality across the platform
Access to cutting-edge language model capabilities
Translation Enhancements
DeepL Integration
Professional-grade translation service option
Expanded translation capabilities beyond Raiqon Translator
Access to well-established DeepL translation quality
Additional choice for translation tasks
Note: Requires DeepL token and incurs additional usage costs
📋 Technical Notes
System Requirements
No changes to existing system requirements
All improvements are backward compatible
Known Limitations
DeepL translation requires separate token and increases usage costs
Model availability is continuously updated
Different model versions may have varying performance characteristics
Configuration Notes
Users can now select from multiple AI models based on their specific needs
Performance optimization depends on model selection
Cost considerations apply when using premium services like DeepL