Mandark-droid

Merge a trained LoRA adapter into the base model, convert to Cactus binary format for mobile deployment, and push to Hugging Face Hub. Use after training completes with /unsloth-jobs-training.

Build debug or release APKs for the LiquidChat Android app. Use when the user wants to compile, build, or install the LiquidChat APK on a device.

Submit a LoRA fine-tuning job to Hugging Face Jobs using Unsloth optimizations on LFM2.5-1.2B-Instruct. Use when the user wants to train or fine-tune a model on HF Jobs cloud GPUs.