Profile Job Results
Jobs
jprylln0g
Results Ready
Name
zipformer_ZipformerEncoder
Target Device
- Dragonwing IQ-9075 EVK
- Qualcomm Linux 1.7
- Qualcomm® Dragonwing™ IQ-9075 | QCS9075
Creator
ai-hub-support@qti.qualcomm.com
Target Model
Input Specs
x: uint16[1, 71, 80]cached_len_0: int32[2, 1]cached_len_1: int32[4, 1]cached_len_2: int32[3, 1]cached_len_3: int32[2, 1]cached_len_4: int32[4, 1]cached_avg_0: uint16[2, 1, 384]cached_avg_1: uint16[4, 1, 384]cached_avg_2: uint16[3, 1, 384]cached_avg_3: uint16[2, 1, 384]cached_avg_4: uint16[4, 1, 384]cached_key_0: uint16[2, 128, 1, 192]cached_key_1: uint16[4, 64, 1, 192]cached_key_2: uint16[3, 32, 1, 192]cached_key_3: uint16[2, 16, 1, 192]cached_key_4: uint16[4, 64, 1, 192]cached_val_0: uint16[2, 128, 1, 96]cached_val_1: uint16[4, 64, 1, 96]cached_val_2: uint16[3, 32, 1, 96]cached_val_3: uint16[2, 16, 1, 96]cached_val_4: uint16[4, 64, 1, 96]cached_val2_0: uint16[2, 128, 1, 96]cached_val2_1: uint16[4, 64, 1, 96]cached_val2_2: uint16[3, 32, 1, 96]cached_val2_3: uint16[2, 16, 1, 96]cached_val2_4: uint16[4, 64, 1, 96]cached_conv1_0: uint16[2, 1, 384, 30]cached_conv1_1: uint16[4, 1, 384, 30]cached_conv1_2: uint16[3, 1, 384, 30]cached_conv1_3: uint16[2, 1, 384, 30]cached_conv1_4: uint16[4, 1, 384, 30]cached_conv2_0: uint16[2, 1, 384, 30]cached_conv2_1: uint16[4, 1, 384, 30]cached_conv2_2: uint16[3, 1, 384, 30]cached_conv2_3: uint16[2, 1, 384, 30]cached_conv2_4: uint16[4, 1, 384, 30]Completion Time
2/24/2026, 7:01:59 AM
Options
--qairt_version 2.42 --max_profiler_iterations 10Versions
- ONNX Runtime: 1.24.1
- QAIRT: v2.42.0.251225135753_193295
- Qualcomm Linux: 1.7-ver.1.1
- AI Hub: aihub-2026.02.14.0
Estimated Inference Time
18.6 ms
Estimated Peak Memory Usage
4 ‑ 7 MB
Compute Units
NPU
2601
| Stage | Time | Memory |
|---|---|---|
First App Load | 284 ms | 161‑162 MB |
Subsequent App Load | 291 ms | 159‑160 MB |
Inference | 18.6 ms | 4‑7 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 | "2" |
| 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