Instance Segmentation

Performance

Device	Model	Input Size	Number of Classes	Single Thread Latency(ms)	Single Thread FPS	12 Threads Latency(ms)	12 Threads FPS	CPU Latency(ms)	Number of Parameters(M)	FLOPs(B)	GitHub Repository
S600	YOLO11l Seg	640×640	80	4.268	233.082	12.015	967.109	5.0	27.6 M	142.2 M	GitHub
S600	YOLO11m Seg	640×640	80	3.624	274.240	9.977	1161.123	5.0	22.4 M	123.3 M	GitHub
S600	YOLO11n Seg	640×640	80	0.959	1024.863	2.379	4586.104	5.0	2.9 M	10.4 M	GitHub
S600	YOLO11s Seg	640×640	80	1.513	651.042	3.845	2908.329	5.0	10.1 M	35.5 M	GitHub
S600	YOLO11x Seg	640×640	80	8.831	112.923	25.480	459.263	5.0	62.1 M	319.0 M	GitHub
S600	YOLOv8l Seg	640×640	80	5.585	178.335	15.661	743.323	5.0	46.0 M	220.5 M	GitHub
S600	YOLOv8m Seg	640×640	80	3.298	301.000	8.931	1296.706	5.0	27.3 M	100.2 M	GitHub
S600	YOLOv8n Seg	640×640	80	0.940	1038.438	2.236	4790.304	5.0	3.4 M	12.6 M	GitHub
S600	YOLOv8s Seg	640×640	80	1.540	642.329	3.883	2873.027	5.0	11.8 M	42.6 M	GitHub
S600	YOLOv8x Seg	640×640	80	8.884	112.265	25.517	458.336	5.0	71.8 M	344.1 M	GitHub
S600	YOLO26n Seg	640x640	80	1.02	966.75	2.54	4353.03	-	3.13	10.5	GitHub
S600	YOLO26s Seg	640x640	80	1.70	579.68	4.38	2585.38	-	11.51	37.4	GitHub
S600	YOLO26m Seg	640x640	80	3.91	254.03	10.87	1069.46	-	27.11	132.5	GitHub
S600	YOLO26l Seg	640x640	80	4.58	217.11	13.00	895.80	-	31.52	150.9	GitHub
S600	YOLO26x Seg	640x640	80	9.32	107.13	26.92	434.77	-	70.69	337.7	GitHub

Device	Model	bbox-all FP32 / BPU	bbox-small FP32 / BPU	bbox-medium FP32 / BPU	bbox-large FP32 / BPU
S600	YOLO11s Seg	0.394 / 0.383 (97.2 %)	0.184 / 0.162 (88.0 %)	0.442 / 0.434 (98.2 %)	0.582 / 0.581 (99.8 %)
S600	YOLO11m Seg	0.443 / 0.427 (96.4 %)	0.246 / 0.224 (91.1 %)	0.497 / 0.484 (97.4 %)	0.627 / 0.607 (96.8 %)
S600	YOLO11l Seg	0.460 / 0.443 (96.3 %)	0.267 / 0.236 (88.4 %)	0.520 / 0.508 (97.7 %)	0.638 / 0.614 (96.2 %)
S600	YOLO11x Seg	0.474 / 0.456 (96.2 %)	0.283 / 0.244 (86.2 %)	0.529 / 0.515 (97.4 %)	0.652 / 0.636 (97.5 %)
S600	YOLOv8s Seg	0.386 / 0.375 (97.2 %)	0.180 / 0.162 (90.0 %)	0.432 / 0.419 (97.0 %)	0.564 / 0.560 (99.3 %)
S600	YOLOv8m Seg	0.431 / 0.420 (97.4 %)	0.228 / 0.209 (91.7 %)	0.486 / 0.478 (98.4 %)	0.608 / 0.605 (99.5 %)
S600	YOLOv8l Seg	0.453 / 0.435 (96.0 %)	0.258 / 0.223 (86.4 %)	0.502 / 0.496 (98.8 %)	0.626 / 0.611 (97.6 %)
S600	YOLOv8x Seg	0.465 / 0.454 (97.6 %)	0.268 / 0.236 (88.1 %)	0.520 / 0.514 (98.8 %)	0.641 / 0.633 (98.8 %)

Device	Model	mask-all FP32 / BPU	mask-small FP32 / BPU	mask-medium FP32 / BPU	mask-large FP32 / BPU
S600	YOLO11s Seg	0.311 / 0.295 (94.9 %)	0.099 / 0.088 (88.9 %)	0.350 / 0.327 (93.4 %)	0.509 / 0.486 (95.5 %)
S600	YOLO11m Seg	0.347 / 0.322 (92.8 %)	0.136 / 0.123 (90.4 %)	0.396 / 0.362 (91.4 %)	0.549 / 0.506 (92.2 %)
S600	YOLO11l Seg	0.357 / 0.331 (92.7 %)	0.143 / 0.126 (88.1 %)	0.409 / 0.377 (92.2 %)	0.560 / 0.511 (91.3 %)
S600	YOLO11x Seg	0.366 / 0.339 (92.6 %)	0.149 / 0.124 (83.2 %)	0.420 / 0.384 (91.4 %)	0.572 / 0.529 (92.5 %)
S600	YOLOv8s Seg	0.305 / 0.290 (95.1 %)	0.096 / 0.085 (88.5 %)	0.343 / 0.317 (92.4 %)	0.496 / 0.476 (96.0 %)
S600	YOLOv8m Seg	0.337 / 0.319 (94.7 %)	0.121 / 0.109 (90.1 %)	0.386 / 0.360 (93.3 %)	0.533 / 0.506 (94.9 %)
S600	YOLOv8l Seg	0.351 / 0.331 (94.3 %)	0.137 / 0.119 (86.9 %)	0.398 / 0.374 (94.0 %)	0.550 / 0.516 (93.8 %)
S600	YOLOv8x Seg	0.358 / 0.338 (94.4 %)	0.139 / 0.122 (87.8 %)	0.409 / 0.382 (93.4 %)	0.562 / 0.529 (94.1 %)
S600	YOLO26n Seg	- / 0.255	- / 0.061	- / 0.274	- / 0.422
S600	YOLO26s Seg	- / 0.331	- / 0.121	- / 0.370	- / 0.508
S600	YOLO26m Seg	- / 0.358	- / 0.156	- / 0.398	- / 0.538
S600	YOLO26l Seg	- / 0.371	- / 0.158	- / 0.417	- / 0.559
S600	YOLO26x Seg	- / 0.382	- / 0.171	- / 0.429	- / 0.571