juli 164bd2130a Switch model to Q6_K (~1.5 GB), lighter/faster than Q8 ...

- ModelDownloader now pulls the Q6_K quant
- README: reflect Q6_K sizes + note prebuilt APK on Releases

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

2026-08-09 22:12:48 +02:00

..

src/main

Switch model to Q6_K (~1.5 GB), lighter/faster than Q8

2026-08-09 22:12:48 +02:00

build.gradle.kts

On-device fixes: optimized native build, 16 KB align, disable thinking

2026-08-09 22:06:30 +02:00

proguard-rules.pro

Scaffold tsjetpiti: on-device Qwen3.5-2B chat app

2026-08-09 20:24:14 +02:00