Effective Date: October 12, 2026
LLM Connect is designed with a privacy-first approach. It supports both local offline model execution and user-initiated cloud API connections.
Local & Offline Mode: Core offline functionality including running local large language models (GGUF, LiteRT), image attachment processing with vision models, and local model benchmarking takes place entirely on your device without requiring network connectivity.
Cloud API Integrations: If you choose to connect to external cloud providers (such as OpenAI, Anthropic Claude, Google Gemini, Groq, xAI Grok, or Ollama servers), your prompts and attachments are transmitted directly to the respective API endpoints you configure. We do not host intermediate proxy servers, nor do we intercept, store, or monitor your API transmissions.
Encrypted On-Device Storage: Your API keys, chat history, settings, benchmark records, and imported model files are stored locally on your device using secure, encrypted storage mechanisms. They are never uploaded, transmitted, or stored on external servers managed by us.
LLM Connect operates on a "Bring Your Own Model" (BYOM) and "Bring Your Own Key" (BYOK) basis for local models and cloud providers, respectively.
No Responsibility for External Models & Services: The developer of LLM Connect does not supply, host, curate, or monitor any AI models loaded locally or third-party AI APIs configured by the user.
Liability Exemption: We explicitly disclaim all liability and legal responsibility for any outputs, errors, biases, offensive content, hallucination, or consequences arising from the specific models you choose to import, or third-party API services you interact with within the application.
Current Data Collection: In its current version, LLM Connect does not collect, track, or sell any personal data, user identifiers, prompt inputs, saved history, or model preferences.
Future Analytical Data: We reserve the right to introduce non-personally identifiable, anonymized analytical data collection in future updates (such as crash logs, hardware performance metrics, or feature usage frequency) strictly to improve app stability and hardware optimizations. Any such telemetry will remain completely anonymous.
LLM Connect is not intended for children (individuals under the age of 13, or equivalent minimum age in your jurisdiction). We do not knowingly target, solicit, or collect personal information from children. If you are under the required age, you are not authorized to use this application.
LLM Connect is proprietary, closed-source software. All rights, title, and interest in and to the application code, architecture, user interface design, and trademarks belong exclusively to the developer. Reverse engineering, decompiling, disassembling, or unauthorized redistribution of the application binaries is strictly prohibited.
We may update this Privacy Policy from time to time to reflect updates in functionality or future feature releases. Any revisions will be published within the application or on its distribution channel with an updated effective date.
If you have any questions, concerns, or inquiries regarding this Privacy Policy, please contact:
Developer Contact: Michael Odonkor
Email: michaelodonkor8@gmail.com