OPPO Open-Sources X-OmniClaw — Edge-Native Multimodal Android Agent Combines Camera, Screen, and Voice On-Device

GitHub·medium signal

OPPO's Multi-X team released X-OmniClaw, an open-source AI agent that runs directly on physical Android devices, combining camera, screen, and voice perception with local memory and autonomous action. The agent processes gallery photos locally into searchable text-based memory, learns by cloning user behavior, and uses a minimal Observation → Reasoning → Execution loop for device manipulation. A cloud LLM is only called as 'fuel' for higher-level reasoning when needed. Accompanied by an arXiv technical report (2605.05765).