Qualcomm® AI HubAI Hub
Qualcomm® AI Hub Models
Track-Anything

Profile Job Results

Jobs
jg98wmdvp
Results Ready
Name
track_anything_segment
Target Device
  • QCS8550 (Proxy)
  • Android 12
  • Qualcomm® QCS8550
Creator
ai-hub-support@qti.qualcomm.com
Input Specs
f16: float32[1, 1024, 20, 36]
f8: float32[1, 512, 40, 72]
f4: float32[1, 256, 80, 144]
memory_readout: float32[1, 1, 512, 20, 36]
hidden_state: float32[1, 1, 64, 20, 36]
Completion Time
6/1/2026, 5:41:56 PM
Versions
  • LiteRT: 1.4.3
  • QAIRT: v2.45.0.260326154327
  • QNN TfLite Delegate: v2.45.0.260326154327
  • Android: 14 (UP1A.231005.007)
  • AI Hub: aihub-2026.05.27.0
Estimated Inference Time
23.4 ms
Estimated Peak Memory Usage
2 ‑ 5 MB
Compute Units
NPU
120
StageTimeMemory
First App Load
3.52 s260‑262 MB
Subsequent App Load
282 ms15‑127 MB
Inference
23.4 ms2‑5 MB
Lite RTValue
number_of_threads4
QNN DelegateValue
backend_typekHtpBackend
log_levelkLogLevelWarn
htp_options.performance_modekHtpBurst
htp_options.precisionkHtpFp16
htp_options.optimization_strategykHtpOptimizeForInferenceO3
htp_options.useConvHmxtrue

Sign up to run this model on a hosted Qualcomm® device!

Run on device