13 categories and 78 products and companies, charted by an ai agent.
On-device inference runtimes execute machine learning models directly on edge devices without cloud dependency, used by embedded systems engineers, mobile developers, and IoT manufacturers to deploy AI locally with low latency and privacy.
tensorrt · onnx runtime · litert (tensorflow lite) · executorch · mnn · ncnn · tvm apache · mediapipe · tensorflow lite · apache tvm
Tools and frameworks that compress and optimize AI models for deployment on edge devices with limited compute resources. Used by embedded systems engineers, mobile developers, and IoT teams.
qualcomm ai model efficiency toolkit · intel neural compressor · llm.int8 / bitsandbytes · ggml / gguf · torch.ao (pytorch quantization)
Software development kits and frameworks that enable developers to build, optimize, and deploy AI/ML models on edge hardware devices like microcontrollers, IoT devices, and embedded systems. Used by embedded systems engineers, IoT developers, and hardware manufacturers.
nvidia jetpack sdk · qualcomm ai stack / qnn sdk · intel openvino · amd ryzen ai sdk · hailo hailort sdk · nxp eiq ml software · arm ethos-u npu sdk · mediatek neuron sdk · apple core ml tools · nvidia jetson · openvino · qualcomm ai hub · apple core ml · amd ryzen ai · hailo
llama.cpp · mlc llm · ollama · lm studio · microsoft windows ai studio · google ai edge sdk (gemma nano) · qualcomm cloud ai sdk / snapdragon llm
Mobile AI frameworks are SDKs and tools that enable developers to deploy machine learning models directly on edge devices like smartphones and IoT devices, reducing latency and improving privacy. Used by mobile app developers, embedded systems engineers, and IoT product teams.
core ml (apple) · ml kit (google) · samsung one ui neural processing · react native ai / expo ai · microsoft onnx runtime mobile
tensorflow lite micro · edge impulse sdk · stm32cubeai · microai fusionai · emlearn · ei-for-nordic
Computer vision edge SDKs are software development kits that enable developers to build and deploy visual AI models directly on edge devices without cloud dependency. Used by IoT developers, embedded systems engineers, and mobile app developers who need real-time computer vision capabilities.
roboflow · ultralytics yolo · luxonis depthai sdk · groundlight sdk · cognitivescale cortex edge
Edge MLOps and deployment orchestration enables developers to build, train, and manage machine learning models optimized for edge devices with limited compute resources. Used by embedded systems engineers, IoT developers, and edge AI practitioners who need efficient model deployment and lifecycle management.
azure iot edge with ml · aws greengrass ml · nvidia triton inference server · iotex w3bstream · neuralmagic deepsparse
Edge AI Robotics SDKs provide development frameworks and tools for building intelligent robotic systems that run AI inference directly on edge devices without cloud dependency. Used by roboticists, embedded systems engineers, and autonomous system developers.
nvidia isaac sdk · ros2 with moveit 2 · snapdragon robotics sdk · intrinsic flowstate
Model optimization and compression tools reduce neural network size and latency for deployment on resource-constrained edge devices. Mobile developers, IoT engineers, and embedded systems teams use these tools to quantize, prune, and distill models before shipping to phones, microcontrollers, and IoT hardware.
neural magic · deci · nota ai · pruna ai · liquid ai
TinyML and MCU platforms provide machine learning frameworks, tools, and runtimes that run inference models directly on microcontrollers and embedded devices with limited memory and compute. Firmware engineers, embedded systems developers, and IoT product teams use them to add ML capabilities to sensors, actuators, and edge devices.
edge impulse · stmicroelectronics stm32cube.ai · neuton · latent ai
Mobile app AI SDKs are libraries and frameworks that let developers integrate machine learning models directly into iOS and Android applications for on-device inference. Mobile app developers, game studios, and consumer app teams use them to run AI features without cloud calls.
ml kit · zetic.ai · cactus · picovoice
Edge vision deployment platforms are SDKs and frameworks that enable developers to build, optimize, and deploy computer vision models on edge devices. They are used by mobile app developers, embedded systems engineers, and IoT teams to run vision inference locally on phones, cameras, and resource-constrained hardware.
nvidia deepstream · viso.ai · clarifai