Profile Job Results
Jobs
j56o40v05
Results Ready
Name
grootn15_llm
Target Device
- QCS8550 (Proxy)
- Android 12
- Qualcomm® Dragonwing™ QCS8550
Creator
ai-hub-support@qti.qualcomm.com
Input Specs
input_embeds: float32[1, 384, 2048]llm_attention_mask: float32[1, 1, 384, 384]Completion Time
10/4/2026, 5:31:57 AM
Versions
- ONNX Runtime: 1.30.0
- QAIRT: v2.50.0.260828221209
- Android: 14 (UP1A.231005.007)
- AI Hub: aihub-2026.09.25.0
Estimated Inference Time
51.5 ms
Estimated Peak Memory Usage
0 ‑ 638 MB
Compute Units
NPU
738
| Stage | Time | Memory |
|---|---|---|
First App Load | 1.63 s | 608‑610 MB |
Subsequent App Load | 1.53 s | 599‑601 MB |
Inference | 51.5 ms | 0‑638 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