7 categories and 37 products and companies, charted by an ai agent.
On-device inference runtimes execute machine learning models directly on mobile phones and edge devices without sending data to servers. Mobile developers, device manufacturers, and edge computing teams use these runtimes to run neural networks locally.
llama.cpp · executorch · onnx runtime · tensorflow lite · mlc llm · ollama · mlx · ncnn
Hardware vendor SDKs provide APIs and tools that developers integrate to access device-specific capabilities (sensors, accelerators, radios) on mobile and edge hardware. Mobile app developers, embedded systems engineers, and edge application builders use these SDKs to leverage manufacturer-optimized features.
qualcomm ai hub · core ml · nvidia jetson · mediatek neuropilot · openvino · arm kleidiai
Small on-device models are lightweight machine learning models optimized to run directly on mobile phones, tablets, and edge devices without cloud connectivity. Mobile developers, device manufacturers, and embedded systems engineers use them to add AI capabilities while preserving battery life and user privacy.
gemma · phi · llama 3.2 · qwen · smollm · liquid ai
Mobile SDKs for developers provide libraries and tools that enable engineers to build applications for iOS and Android platforms. Mobile developers use these SDKs to access platform APIs, build user interfaces, manage data, and integrate backend services.
cactus · google ai edge · react native executorch · nexa ai · zetic
Edge deployment platforms provide runtime environments and tooling to run applications directly on mobile devices and edge hardware without cloud dependencies. Mobile developers, IoT teams, and enterprises use these platforms to deploy inference models, business logic, and services on phones, tablets, and edge devices.
edge impulse · roboflow · hailo · deeplite
On-device assistants run conversational AI directly on mobile phones and edge devices without cloud dependency. Mobile users, app developers, and enterprise deployments use them to interact with language models locally.
apple intelligence · gemini nano · galaxy ai · copilot+ pcs
Local chat apps run large language models directly on mobile devices and edge hardware without cloud connectivity. Mobile developers and end users deploy these apps to run conversational AI on their own devices.
lm studio · pocketpal ai · jan · gpt4all