Smart routing between on-device and local GPU inference
| Condition | Model | Latency |
|---|---|---|
| WiFi + Laptop available | Gemma 4 E4B via Ollama | < 1s |
| No WiFi, model installed | Gemma 4 E2B on-device | 3-8s |
| WiFi + No laptop | Gemma 4 E2B on-device | 3-8s |
| No WiFi, no model | Prompt to download | โ |
Gemma 4 E2B / E4B
LiteRT-LM (on-device) + Ollama (GPU)
Flutter 3.41 / Dart
SM-2 Spaced Repetition
ML Kit (offline) + Gemma 4 multimodal
SharedPreferences + JSON
Your data never leaves your device. No cloud. No tracking. No API keys. Open source under Apache 2.0.