RunAnywhere

RunAnywhere lets you run capable models directly on supported Android devices.

Chat with downloaded language models, analyze photos and live camera input, ask questions across documents with cited source chunks, and use speech-to-text and text-to-speech from one app.

Features include:

• Local chat with device-aware model recommendations
• Image and chart understanding
• Live camera analysis
• Document question answering with cited sources
• Hands-free voice conversations
• Speech transcription and voice generation
• Built-in tools for tasks such as calculations
• Optional web search with visible tool traces and source URLs
• Qualcomm Hexagon NPU acceleration on supported devices
• Model download and storage controls

Core inference runs on your device after the required model files are downloaded. Network access is required for model downloads. Optional connected features can use network services when enabled. Feature availability and performance depend on the selected model and device hardware.