Skip to main content

What's New in Version 0.629.0

Released on September 30, 2025

We're excited to bring you significant improvements to our AI model reasoning capabilities and enhanced orchestration features! This release introduces powerful reasoning boosts, advanced agent collaboration, and intelligent knowledge retrieval to make your AI workflows more effective than ever.

🧠 Enhanced AI Reasoning

Boosted Reasoning Capabilities

Both Sorcerer and Mystica models have received major reasoning improvements, delivering more accurate and sophisticated responses across all tasks. Experience enhanced problem-solving and more nuanced understanding in your AI interactions.

🤖 Advanced Agent Orchestration

Orchestrator-Operator Collaboration with Agentic Tooling

Orchestrator agents now seamlessly collaborate with Operator agents leveraging Agentic Tooling, creating streamlined planning workflows while maintaining human-in-the-loop approval for critical checkpoints and smart notifications. This powerful combination ensures both efficiency and control in your agent workflows.

Enhanced Dynamic Agent Creation

Dynamic agent creation now leverages the latest DeepSeek 3.1 Terminus alongside Sorcerer and Mystica models for greater flexibility and accuracy. By default, when Orchestrator agents spawn new agents, you'll see Operators with Agentic Tooling and Assistants. Note that legacy Coder and Operator agents can still be statically assigned to orchestrators for specific use cases.

🎯 Intelligent Knowledge Management

Dynamic Topic Filtering

Operators now feature intelligent topic filtering that uses topic descriptions to optimise knowledge retrieval. This enhancement ensures more relevant and accurate information retrieval from your knowledge hub, improving the overall effectiveness of your AI agents.

🛠️ Improvements & Fixes

Model Reasoning Fixes

  • Fixed an issue where Mystica 1.5 thinking would not reason under certain conditions, ensuring consistent cognitive processing
  • Resolved DeepSeek 3.1 Terminus issue where thinking tags weren't properly applied, allowing the full reasoning to display in core answers
  • Corrected image labeling during the learning process, which was preventing proper image retrieval using RAG

📋 Summary

This update includes:

  • 4 major new features that enhance reasoning and orchestration capabilities
  • 3 critical bug fixes for improved model performance and reliability

Thank you for using ToothFairy! We're constantly working to improve your experience. If you have any feedback or encounter any issues, please don't hesitate to reach out to our support team at support@toothfairyai.com.

Keep building magic with AI! 🧚‍♀️✨

v0.631.1