VajraV1 -- The most accurate Real Time Object Detector of the YOLO family

Benchmark Model Rank Results
real-time-object-detection-on-coco-val2017VajraV1-Xlarge#13box AP: 56.2FPS (batch 1, GPU in brackets): 312 (RTX 4090)
real-time-object-detection-on-coco-val2017VajraV1-Large#35box AP: 53.7FPS (batch 1, GPU in brackets): 556 (RTX 4090)
real-time-object-detection-on-coco-val2017VajraV1-Medium#55box AP: 52.7FPS (batch 1, GPU in brackets): 667 (RTX 4090)
real-time-object-detection-on-coco-val2017VajraV1-Small#78box AP: 50.4FPS (batch 1, GPU in brackets): 909 (RTX 4090)
real-time-object-detection-on-coco-val2017VajraV1-Nano#113box AP: 44.3FPS (batch 1, GPU in brackets): 909 (RTX 4090)