验证tensorRT10.0 trt模型是否有效

trtexec --loadEngine=model.trt
rtx3060显卡显示pass则OK

&&&& PASSED TensorRT.trtexec [TensorRT v100001] # trtexec --loadEngine=model.trt

D:\cdtxw\TensorRT-10.0.1.6.Windows10.win10.cuda-12.4\TensorRT-10.0.1.6\bin>trtexec --loadEngine=model.trt
&&&& RUNNING TensorRT.trtexec [TensorRT v100001] # trtexec --loadEngine=model.trt
[04/29/2026-13:23:53] [I] === Model Options ===
[04/29/2026-13:23:53] [I] Format: *
[04/29/2026-13:23:53] [I] Model:
[04/29/2026-13:23:53] [I] Output:
[04/29/2026-13:23:53] [I]
[04/29/2026-13:23:53] [I] === System Options ===
[04/29/2026-13:23:53] [I] Device: 0
[04/29/2026-13:23:53] [I] DLACore:
[04/29/2026-13:23:53] [I] Plugins:
[04/29/2026-13:23:53] [I] setPluginsToSerialize:
[04/29/2026-13:23:53] [I] dynamicPlugins:
[04/29/2026-13:23:53] [I] ignoreParsedPluginLibs: 0
[04/29/2026-13:23:53] [I]
[04/29/2026-13:23:53] [I] === Inference Options ===
[04/29/2026-13:23:53] [I] Batch: Explicit
[04/29/2026-13:23:53] [I] Input inference shapes: model
[04/29/2026-13:23:53] [I] Iterations: 10
[04/29/2026-13:23:53] [I] Duration: 3s (+ 200ms warm up)
[04/29/2026-13:23:53] [I] Sleep time: 0ms
[04/29/2026-13:23:53] [I] Idle time: 0ms
[04/29/2026-13:23:53] [I] Inference Streams: 1
[04/29/2026-13:23:53] [I] ExposeDMA: Disabled
[04/29/2026-13:23:53] [I] Data transfers: Enabled
[04/29/2026-13:23:53] [I] Spin-wait: Disabled
[04/29/2026-13:23:53] [I] Multithreading: Disabled
[04/29/2026-13:23:53] [I] CUDA Graph: Disabled
[04/29/2026-13:23:53] [I] Separate profiling: Disabled
[04/29/2026-13:23:53] [I] Time Deserialize: Disabled
[04/29/2026-13:23:53] [I] Time Refit: Disabled
[04/29/2026-13:23:53] [I] NVTX verbosity: 0
[04/29/2026-13:23:53] [I] Persistent Cache Ratio: 0
[04/29/2026-13:23:53] [I] Optimization Profile Index: 0
[04/29/2026-13:23:53] [I] Weight Streaming Budget: Disabled
[04/29/2026-13:23:53] [I] Inputs:
[04/29/2026-13:23:53] [I] Debug Tensor Save Destinations:
[04/29/2026-13:23:53] [I] === Reporting Options ===
[04/29/2026-13:23:53] [I] Verbose: Disabled
[04/29/2026-13:23:53] [I] Averages: 10 inferences
[04/29/2026-13:23:53] [I] Percentiles: 90,95,99
[04/29/2026-13:23:53] [I] Dump refittable layers:Disabled
[04/29/2026-13:23:53] [I] Dump output: Disabled
[04/29/2026-13:23:53] [I] Profile: Disabled
[04/29/2026-13:23:53] [I] Export timing to JSON file:
[04/29/2026-13:23:53] [I] Export output to JSON file:
[04/29/2026-13:23:53] [I] Export profile to JSON file:
[04/29/2026-13:23:53] [I]
[04/29/2026-13:23:53] [I] === Device Information ===
[04/29/2026-13:23:53] [I] Available Devices:
[04/29/2026-13:23:53] [I]   Device 0: "NVIDIA GeForce RTX 3060" UUID: GPU-fdec3edd-55e9-18e8-70ae-dd99ad5ef70b
[04/29/2026-13:23:54] [I] Selected Device: NVIDIA GeForce RTX 3060
[04/29/2026-13:23:54] [I] Selected Device ID: 0
[04/29/2026-13:23:54] [I] Selected Device UUID: GPU-fdec3edd-55e9-18e8-70ae-dd99ad5ef70b
[04/29/2026-13:23:54] [I] Compute Capability: 8.6
[04/29/2026-13:23:54] [I] SMs: 28
[04/29/2026-13:23:54] [I] Device Global Memory: 12287 MiB
[04/29/2026-13:23:54] [I] Shared Memory per SM: 100 KiB
[04/29/2026-13:23:54] [I] Memory Bus Width: 192 bits (ECC disabled)
[04/29/2026-13:23:54] [I] Application Compute Clock Rate: 1.777 GHz
[04/29/2026-13:23:54] [I] Application Memory Clock Rate: 7.501 GHz
[04/29/2026-13:23:54] [I]
[04/29/2026-13:23:54] [I] Note: The application clock rates do not reflect the actual clock rates that the GPU is currently running at.
[04/29/2026-13:23:54] [I]
[04/29/2026-13:23:54] [I] TensorRT version: 10.0.1
[04/29/2026-13:23:54] [I] Loading standard plugins
[04/29/2026-13:23:54] [I] [TRT] Loaded engine size: 31 MiB
[04/29/2026-13:23:54] [I] Engine deserialized in 0.0506413 sec.
[04/29/2026-13:23:54] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in IExecutionContext creation: CPU +0, GPU +32, now: CPU 0, GPU 59 (MiB)
[04/29/2026-13:23:54] [I] Setting persistentCacheLimit to 0 bytes.
[04/29/2026-13:23:54] [I] Created execution context with device memory size: 32.5684 MiB
[04/29/2026-13:23:54] [I] Using random values for input images
[04/29/2026-13:23:54] [I] Input binding for images with dimensions 1x3x640x640 is created.
[04/29/2026-13:23:54] [I] Output binding for output0 with dimensions 1x25200x6 is created.
[04/29/2026-13:23:54] [I] Starting inference
[04/29/2026-13:23:57] [I] Warmup completed 55 queries over 200 ms
[04/29/2026-13:23:57] [I] Timing trace has 912 queries over 3.00798 s
[04/29/2026-13:23:57] [I]
[04/29/2026-13:23:57] [I] === Trace details ===
[04/29/2026-13:23:57] [I] Trace averages of 10 runs:
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.27158 ms - Host latency: 3.55169 ms (enqueue 0.235562 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.26227 ms - Host latency: 3.54421 ms (enqueue 0.229588 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.25232 ms - Host latency: 3.5312 ms (enqueue 0.233936 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.26164 ms - Host latency: 3.55309 ms (enqueue 0.312955 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.27527 ms - Host latency: 3.55285 ms (enqueue 0.263553 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.24512 ms - Host latency: 3.52156 ms (enqueue 0.234555 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.27122 ms - Host latency: 3.54536 ms (enqueue 0.279669 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.25597 ms - Host latency: 3.52809 ms (enqueue 0.232373 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.25819 ms - Host latency: 3.54006 ms (enqueue 0.236603 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.25058 ms - Host latency: 3.52447 ms (enqueue 0.225491 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.25029 ms - Host latency: 3.52529 ms (enqueue 0.351422 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.26315 ms - Host latency: 3.54017 ms (enqueue 0.248328 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.25595 ms - Host latency: 3.53047 ms (enqueue 0.225879 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.26296 ms - Host latency: 3.53982 ms (enqueue 0.229376 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.26677 ms - Host latency: 3.54118 ms (enqueue 0.227271 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.26885 ms - Host latency: 3.54128 ms (enqueue 0.226233 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.33024 ms - Host latency: 3.60322 ms (enqueue 0.274139 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.25842 ms - Host latency: 3.5301 ms (enqueue 0.23075 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.24912 ms - Host latency: 3.52295 ms (enqueue 0.226483 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.25103 ms - Host latency: 3.52394 ms (enqueue 0.35061 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.25674 ms - Host latency: 3.52866 ms (enqueue 0.252045 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.26942 ms - Host latency: 3.54549 ms (enqueue 0.268597 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.25859 ms - Host latency: 3.5386 ms (enqueue 0.2755 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.25273 ms - Host latency: 3.52945 ms (enqueue 0.226239 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.25125 ms - Host latency: 3.52156 ms (enqueue 0.230713 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.24615 ms - Host latency: 3.52579 ms (enqueue 0.241553 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.26956 ms - Host latency: 3.54294 ms (enqueue 0.320264 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.24769 ms - Host latency: 3.53057 ms (enqueue 0.424158 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.24855 ms - Host latency: 3.52181 ms (enqueue 0.227209 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.25098 ms - Host latency: 3.53075 ms (enqueue 0.232568 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.24949 ms - Host latency: 3.52338 ms (enqueue 0.228137 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.26489 ms - Host latency: 3.5382 ms (enqueue 0.224622 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.33331 ms - Host latency: 3.61204 ms (enqueue 0.241907 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.25673 ms - Host latency: 3.53912 ms (enqueue 0.275366 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.25276 ms - Host latency: 3.52327 ms (enqueue 0.234912 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.25162 ms - Host latency: 3.52848 ms (enqueue 0.25531 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.23345 ms - Host latency: 3.50366 ms (enqueue 0.226343 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.26809 ms - Host latency: 3.54641 ms (enqueue 0.310645 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.26123 ms - Host latency: 3.53636 ms (enqueue 0.225623 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.24888 ms - Host latency: 3.52233 ms (enqueue 0.227661 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.25553 ms - Host latency: 3.52748 ms (enqueue 0.310608 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.27717 ms - Host latency: 3.54877 ms (enqueue 0.486438 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.25909 ms - Host latency: 3.51829 ms (enqueue 0.236597 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.26716 ms - Host latency: 3.53732 ms (enqueue 0.233606 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.27636 ms - Host latency: 3.5307 ms (enqueue 0.229749 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.2725 ms - Host latency: 3.54008 ms (enqueue 0.230872 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.29437 ms - Host latency: 3.57953 ms (enqueue 0.229053 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.31082 ms - Host latency: 3.59656 ms (enqueue 0.225122 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.34286 ms - Host latency: 3.61547 ms (enqueue 0.267114 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.25658 ms - Host latency: 3.52594 ms (enqueue 0.226233 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.25605 ms - Host latency: 3.5321 ms (enqueue 0.230554 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.26814 ms - Host latency: 3.5452 ms (enqueue 0.265808 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.25149 ms - Host latency: 3.52341 ms (enqueue 0.338147 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.26481 ms - Host latency: 3.54089 ms (enqueue 0.24082 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.27865 ms - Host latency: 3.55026 ms (enqueue 0.227759 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.26298 ms - Host latency: 3.53359 ms (enqueue 0.231311 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.26667 ms - Host latency: 3.54 ms (enqueue 0.317224 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.27161 ms - Host latency: 3.54849 ms (enqueue 0.309863 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.25444 ms - Host latency: 3.5197 ms (enqueue 0.236108 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.27859 ms - Host latency: 3.55837 ms (enqueue 0.233203 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.28555 ms - Host latency: 3.56121 ms (enqueue 0.266992 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.2625 ms - Host latency: 3.53491 ms (enqueue 0.227295 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.26868 ms - Host latency: 3.54458 ms (enqueue 0.23064 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.28748 ms - Host latency: 3.56067 ms (enqueue 0.22981 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.33057 ms - Host latency: 3.61541 ms (enqueue 0.294409 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.26702 ms - Host latency: 3.5417 ms (enqueue 0.228857 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.27664 ms - Host latency: 3.54985 ms (enqueue 0.225977 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.28008 ms - Host latency: 3.56162 ms (enqueue 0.312085 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.276 ms - Host latency: 3.53518 ms (enqueue 0.320728 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.27627 ms - Host latency: 3.54097 ms (enqueue 0.245654 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.29497 ms - Host latency: 3.5739 ms (enqueue 0.266357 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.28687 ms - Host latency: 3.56318 ms (enqueue 0.347046 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.28574 ms - Host latency: 3.55991 ms (enqueue 0.233521 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.27397 ms - Host latency: 3.5512 ms (enqueue 0.225562 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.27397 ms - Host latency: 3.54617 ms (enqueue 0.246875 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.26941 ms - Host latency: 3.54966 ms (enqueue 0.225635 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.28464 ms - Host latency: 3.55774 ms (enqueue 0.345483 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.28767 ms - Host latency: 3.56484 ms (enqueue 0.22627 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.2917 ms - Host latency: 3.56902 ms (enqueue 0.232324 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.28391 ms - Host latency: 3.56431 ms (enqueue 0.226978 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.33992 ms - Host latency: 3.61333 ms (enqueue 0.259058 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.27871 ms - Host latency: 3.55242 ms (enqueue 0.23147 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.26851 ms - Host latency: 3.54167 ms (enqueue 0.22688 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.28333 ms - Host latency: 3.55967 ms (enqueue 0.316919 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.27964 ms - Host latency: 3.5583 ms (enqueue 0.274243 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.27249 ms - Host latency: 3.54912 ms (enqueue 0.241016 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.29192 ms - Host latency: 3.57085 ms (enqueue 0.226147 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.26965 ms - Host latency: 3.54851 ms (enqueue 0.275024 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.26995 ms - Host latency: 3.55879 ms (enqueue 0.330786 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.27021 ms - Host latency: 3.546 ms (enqueue 0.223511 ms)
[04/29/2026-13:23:57] [I] Average on 10 runs - GPU latency: 3.27686 ms - Host latency: 3.54849 ms (enqueue 0.315283 ms)
[04/29/2026-13:23:57] [I]
[04/29/2026-13:23:57] [I] === Performance summary ===
[04/29/2026-13:23:57] [I] Throughput: 303.194 qps
[04/29/2026-13:23:57] [I] Latency: min = 3.46265 ms, max = 4.35815 ms, mean = 3.54589 ms, median = 3.5403 ms, percentile(90%) = 3.5813 ms, percentile(95%) = 3.59961 ms, percentile(99%) = 3.65942 ms
[04/29/2026-13:23:57] [I] Enqueue Time: min = 0.220703 ms, max = 0.797729 ms, mean = 0.259363 ms, median = 0.226807 ms, percentile(90%) = 0.270996 ms, percentile(95%) = 0.61084 ms, percentile(99%) = 0.673584 ms
[04/29/2026-13:23:57] [I] H2D Latency: min = 0.20752 ms, max = 0.346649 ms, mean = 0.230353 ms, median = 0.230957 ms, percentile(90%) = 0.235107 ms, percentile(95%) = 0.23999 ms, percentile(99%) = 0.266113 ms
[04/29/2026-13:23:57] [I] GPU Compute Time: min = 3.20331 ms, max = 4.08569 ms, mean = 3.27073 ms, median = 3.26453 ms, percentile(90%) = 3.30371 ms, percentile(95%) = 3.31885 ms, percentile(99%) = 3.38477 ms
[04/29/2026-13:23:57] [I] D2H Latency: min = 0.0324707 ms, max = 0.107971 ms, mean = 0.0448103 ms, median = 0.0471191 ms, percentile(90%) = 0.0501099 ms, percentile(95%) = 0.0530396 ms, percentile(99%) = 0.0649414 ms
[04/29/2026-13:23:57] [I] Total Host Walltime: 3.00798 s
[04/29/2026-13:23:57] [I] Total GPU Compute Time: 2.9829 s
[04/29/2026-13:23:57] [W] * GPU compute time is unstable, with coefficient of variance = 1.88254%.
[04/29/2026-13:23:57] [W]   If not already in use, locking GPU clock frequency or adding --useSpinWait may improve the stability.
[04/29/2026-13:23:57] [I] Explanations of the performance metrics are printed in the verbose logs.
[04/29/2026-13:23:57] [I]
&&&& PASSED TensorRT.trtexec [TensorRT v100001] # trtexec --loadEngine=model.trt