Tellama

Tellama is a local LLM chat app for Android users who want conversations to stay on their phone. Chat with an installed compatible model works offline, and you don't need an account.

• Local chat and Memory: Read responses as they're generated and retain user-approved Memory alongside prompts, replies and chat history on the device.
• Model choice: Browse a catalog with guidance on model fit for your phone, use Tellama-verified distributions such as TM Qwen3 4B with Deep thinking, or import compatible GGUF models that need separate validation.
• Device readiness: Check available RAM, storage and model readiness in the dashboard; if a replacement model fails to load, the app restores the previous model and selection.
• Optional network features: Connect to an external API model or use web research, with an explanation of what data will leave the phone before transmission.

Tellama runs on Android 10 and later and distributes its signed APK outside Google Play. It verifies model downloads and app updates, including the update's signing certificate. Downloads require internet access, as do external API models and web research.

External models receive your input and required context under the selected provider's terms. Web research connects to the public HTTPS URL you choose; if you use an external model for summarization, that provider also receives the page content. Performance depends on the phone and model, and Deep thinking can take minutes before a visible reply appears.