Ever wished you could just ask your simulation why a packet was dropped, or let an AI co-pilot diagnose a network anomaly for you?

We are excited to share the OMNeT++ 6.4.0 AI & LLM Technology Preview 2, bringing the power of Large Language Models directly into your simulation workflow!

We need your help to test these experimental features and shape the future of AI-assisted simulation. Here is what you can do with this new preview:

What’s Inside?

• Chat directly with Qtenv: A new integrated chat interface in Qtenv lets you interact with LLM services (both cloud-based and local models like Ollama) without leaving the simulation window.
• AI-Controlled Simulations: Built on the standard Model Context Protocol (MCP), the simulation hosts an MCP server. This allows LLMs to query and programmatically manipulate your active simulations in real time.
• On-the-Fly C++ Execution (Linux-only): The LLM can generate and run custom C++ snippets using a JIT interpreter to inspect complex data structures and answer highly specific queries. (Launched safely via the opp_sandbox utility).

See It in Action

We have prepared two practical, hands-on showcases to demonstrate these features:

• Finding a 3-Way Collision in Aloha – Watch an AI agent pinpoint a rare collision event.
• Diagnosing Clock Sync Anomalies in TSN – Witness an AI co-pilot perform root-cause analysis on a Time-Sensitive Network.

How to Try It & Give Feedback

This preview is available for Linux (x86_64) and macOS (aarch64).

1. Head over to the Preview Downloads page to download the pre-built packages.
2. Read the Release Notes for detailed setup instructions.
3. Share your thoughts, bugs, and feature ideas with us on GitHub Discussions!

Your feedback is welcome! Please download it, try the showcases, run it against your own models, and let us know what you think!