Profile Job Results
Jobs
j5wlxnnzp
Results Ready
Name
grootn15_dit
Target Device
- Dragonwing IQ-9075 EVK
- Qualcomm Linux 1.9
- Qualcomm® Dragonwing™ IQ-9075 | QCS9075
Creator
ai-hub-support@qti.qualcomm.com
Input Specs
state: float32[1, 1, 64]actions: float32[1, 16, 32]vlm_proj_keys_0: float32[1, 32, 384, 48]vlm_proj_values_0: float32[1, 32, 384, 48]cross_attention_mask: float32[1, 32, 49, 384]vlm_proj_keys_1: float32[1, 32, 384, 48]vlm_proj_values_1: float32[1, 32, 384, 48]vlm_proj_keys_2: float32[1, 32, 384, 48]vlm_proj_values_2: float32[1, 32, 384, 48]vlm_proj_keys_3: float32[1, 32, 384, 48]vlm_proj_values_3: float32[1, 32, 384, 48]vlm_proj_keys_4: float32[1, 32, 384, 48]vlm_proj_values_4: float32[1, 32, 384, 48]vlm_proj_keys_5: float32[1, 32, 384, 48]vlm_proj_values_5: float32[1, 32, 384, 48]vlm_proj_keys_6: float32[1, 32, 384, 48]vlm_proj_values_6: float32[1, 32, 384, 48]vlm_proj_keys_7: float32[1, 32, 384, 48]vlm_proj_values_7: float32[1, 32, 384, 48]Completion Time
9/20/2026, 5:46:12 AM
Options
--qairt_version latestVersions
- ONNX Runtime: 1.27.1
- QAIRT: v2.50.0.260828221209
- Qualcomm Linux: 1.9-ver.1.0
- AI Hub: aihub-2026.09.11.0
Estimated Inference Time
94.5 ms
Estimated Peak Memory Usage
38 ‑ 80 MB
Compute Units
NPU
1880
| Stage | Time | Memory |
|---|---|---|
First App Load | 715 ms | 840‑841 MB |
Subsequent App Load | 734 ms | 838‑839 MB |
Inference | 94.5 ms | 38‑80 MB |
| ONNX Runtime | Value |
|---|---|
| execution_mode | SEQUENTIAL |
| intra_op_num_threads | 0 |
| inter_op_num_threads | 0 |
| enable_memory_pattern | false |
| enable_cpu_memory_arena | false |
| graph_optimization_level | ENABLE_ALL |
| QNN Execution Provider | Value |
|---|---|
| htp_performance_mode | "burst" |
| htp_graph_finalization_optimization_mode | "3" |
| enable_htp_fp16_precision | "1" |
| capture_network_visualizations | false |
| context_priority | "normal" |
| offload_graph_io_quantization | "1" |
Sign up to run this model on a hosted Qualcomm® device!
Run on device