Seeed-Studio/ModelAssistant
Seeed SenseCraft Model Assistant is an open-source project focused on embedded AI. 🔥🔥🔥
Optimized algorithms from OpenMMLab for object detection, image classification, segmentation, pose estimation, and anomaly detection, with support for quantization-aware training (QAT) and model export to TensorFlow Lite, ONNX, TensorRT, and UF2 formats. Includes a complete toolchain: web-based training interface (SenseCraft AI), model zoo, SSCMA-Micro framework for MCU deployment, and Python/Arduino libraries for ESP32, Raspberry Pi, and other resource-constrained devices. RTMDet, YOLOv8, VAE, and Swift YOLO enable real-time inference on hardware consuming under 0.3 watts.
425 stars.
Stars
425
Forks
60
Language
Python
License
—
Category
Last pushed
Feb 02, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Seeed-Studio/ModelAssistant"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
rakesh-i/ESP32-Autonomous-car
Autonomous car using ESP32.
sebastiano123-c/DroneIno
Arduino code for ESP32 based DIY quadcopter drones, providing autonomous flight via artificial...
briand-hub/Esp32AI
C++17 Neural Network (NN), Convolutional Neural Network (CNN) and Deep Learning for Esp32 on IDF...
JarmouniA/TFLM_Inference_ESP32_PlatformIO
Implementing a handwritten digits recognition application on the ESP32-CAM dev board using...
make2explore/M5Stack-AtomS3
Device Overview - Getting Started with M5Stack's AtomS3