Contact & Community

Connect with the maintainers, report issues, or contribute to local-first AI development.

For questions regarding software architecture, research collaborations, partnership opportunities, or feedback on local AI performance.

Email: info@starkllm.com

Responses typically within 24–48 hours on business days.

StarkLLM is completely transparent and hosted on GitHub. Submit bug reports, request features, or review the source code.

Repository: github.com/homomorphicprivacy/StarkLLM

Issue trackers and discussion boards are open for public participation.

Maintained by Christian Stark under the Homomorphic Privacy research initiative.

Address:
Maxstr. 14
52070 Aachen
Germany (Deutschland)

How to Get Assistance

Depending on what you need help with, here are the most effective ways to reach the right resource:

Installation or Setup Issues: Check our detailed User Manual first. It covers Docker configuration, Ollama model setup, GPU acceleration, and troubleshooting port conflicts.
Bug Reports: If you encountered an application crash or unexpected error during document ingestion, please open an issue on GitHub Issues with system specs and reproducible steps.
Responsible Security Disclosures: If you discovered a potential local vulnerability, please email us directly at info@starkllm.com with subject prefix [Security] before public disclosure.
Hardware & Model Sizing: Wondering whether your machine can run Qwen, Mistral, or Llama models comfortably? Review our Local Models Guide or ask in GitHub Discussions.

Yes. StarkLLM is provided under an open-source license. There are no subscriptions, hidden cloud fees, or API paywalls.

Zero. All token generation, vector embeddings, document chunking, and database storage execute locally on your physical machine.

Visit the Download Page to obtain the official Windows Setup installer and verify SHA-256 integrity checksums.