Qualcomm® AI HubAI Hub

Profile Job Results

Jobs
jgzn30kzg
Results Ready
Name
bevdet_pooler
Target Device
  • Dragonwing IQ-9075 EVK
  • Qualcomm Linux 1.7
  • Qualcomm® Dragonwing™ IQ-9075 | QCS9075
Creator
ai-hub-support@qti.qualcomm.com
Input Specs
depth: float32[1, 6, 59, 16, 44]
feat: float32[1, 6, 64, 16, 44]
sensor2keyegos: float32[1, 6, 4, 4]
inv_intrins: float32[1, 6, 3, 3]
inv_post_rots: float32[1, 6, 3, 3]
post_trans: float32[1, 6, 1, 3]
Completion Time
7/25/2026, 9:23:13 PM
Options
--compute_unit cpu
Versions
  • ONNX Runtime: 1.27.1
  • Qualcomm Linux: 1.7-ver.1.1
  • AI Hub: aihub-2026.07.17.0
Estimated Inference Time
56.8 ms
Estimated Peak Memory Usage
330 ‑ 334 MB
Compute Units
CPU
289
StageTimeMemory
First App Load
75.9 ms13‑14 MB
Subsequent App Load
67.7 ms11‑12 MB
Inference
56.8 ms330‑334 MB
ONNX RuntimeValue
execution_modeSEQUENTIAL
intra_op_num_threads0
inter_op_num_threads0
enable_memory_patternfalse
enable_cpu_memory_arenafalse
graph_optimization_levelENABLE_ALL

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

Run on device