# AMV-Train Full Technical Specification (v2.1.0) Official Documentation & Deep Architecture Reference for Autonomous AI Agents. ## 1. System Overview Unified Multimodal On-Device Deep Learning & LoRA Training Framework for Android Termux with 6-Modality Adapters (LLM, Diffusion, VLM, STT, TTS, BitNet), GPU Slicing, and 44GB Disaggregated Cluster Virtual RAM Pooling AMV-Train v2.0.1 delivers a dependency-free C-vectorized DAG Autograd engine, 6-Modality target-aware LoRA/DoRA training (LLM, Image Diffusion, Vision VLM, Whisper STT, TTS, BitNet 1.58-bit), Mali/Adreno GPU Slicing, 300MB Guard-Band memory protection, and 44GB disaggregated AMEVA Cluster virtual RAM pooling across heterogeneous mobile fleets. ## 2. Package & Installation - PyPI: amv-train - Command: pip install amv-train # or via npm: npm install -g amv-train # or standalone bootstrap: bash <(curl -sSL https://raw.githubusercontent.com/uno-km/amv-train/main/install.sh) - Repository: https://github.com/uno-km/amv-train