EdgeLM

Privacy Policy

Effective date: 9 August 2026 · App: EdgeLM Runtime (ai.edgelm.runtime)

EdgeLM runs AI models on your device. It has no user accounts, shows no ads, and does not collect, store, or transmit your personal data. Your prompts, images, recordings and the AI's responses never leave your phone — with one clearly marked exception, spoken replies, which use your device's own text-to-speech engine.

What we collect

Nothing. EdgeLM does not collect, log, or transmit any personal or usage data. There is no analytics SDK, no advertising SDK, no crash-reporting service, and no account system.

How your content is handled

Network use

EdgeLM makes network requests for one purpose only: to download the AI model you explicitly choose from the in-app catalog. These downloads are fetched over HTTPS from third-party model hosts (for example, Hugging Face). Only a standard file request is made — no personal data, prompts, or identifiers are sent by EdgeLM. The host you download from may record standard request metadata (such as your IP address) under its own privacy policy; please refer to that host's policy. If you never download a model, EdgeLM makes no network requests for inference. EdgeLM performs no cloud inference — model execution is always local.

Permissions

EdgeLM requests no access to your location, contacts, camera, messages, or files outside its own private app storage. It has no permission to browse your photos — an image is only ever handed to it one at a time, by you, through the system picker.

Spoken replies (text-to-speech)

If you turn on spoken replies, EdgeLM passes the answer text to your device's own text-to-speech engine (for example, Google Text-to-Speech) — the same one other apps and Android's accessibility features use. This is a system component, not part of EdgeLM, and it is the one feature that may not be fully offline: some engines download or stream voice data, which means the reply text could be handled by that engine's provider under its privacy policy, not ours. If that matters to you, install your TTS engine's offline voice data, or leave spoken replies switched off — they are off unless you enable them, and everything else in EdgeLM continues to run entirely on your device.

Data sharing

EdgeLM does not share any data with third parties, because it does not collect any.

Children

EdgeLM does not knowingly collect data from anyone, including children. It contains no ads and no data collection.

Changes to this policy

If this policy changes, the updated version will be posted at the same URL with a new effective date.

Contact

Questions about this policy: chandramaulisharma26@gmail.com