Unverified50% confidenceBenchmarkExact time
树莓派4B单精度浮点峰值性能约9.6 GFLOPS，可运行OpenCV经典追踪算法但基本无法运行YOLOv5s（约7.2 GFLOPS/帧）
1
Sources
50%
Confidence
Long-term
Relevance
7/15/2026
First Seen
Sources
Related Claims
UnverifiedSince the release of the first version in 2016, the YOLO series has undergone multiple iterations from YOLOv1 to YOLOv10, achieving inference speeds of 100+ FPS.62% similarUnverifiedYOLOv8 模型家族从 YOLOv8n（约 3.2M 参数）到 YOLOv8x（约 68M 参数）覆盖从树莓派到高性能 GPU 的硬件谱系61% similarUnverified相比YOLOv8，YOLOv11在模型架构上引入了C3k2模块和SPPF改进，在保持检测精度的同时降低了参数量和计算量60% similarUnverifiedYOLOv3 introduced multi-scale prediction and residual networks; YOLOv4 integrated CSPNet and PANet.59% similarUnverifiedTensorRT supports multiple precision calibrations including FP32, FP16, INT8, and FP858% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/519418API
curl https://kongchang.com/api/v1/knowledge/claims/519418MCP
get_claim(id=519418)