Profile Job Results
Jobs
jglkxxz2p
Results Ready
Name
melotts_en_T5Decoder_EN
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
input_ids: int32[1, 1]encoder_attention_mask: int32[1, 50]position: int32[1, 1]block_0_past_self_key_states: float32[1, 6, 49, 64]block_0_past_self_value_states: float32[1, 6, 49, 64]block_0_cross_key_states: float32[1, 6, 50, 64]block_0_cross_value_states: float32[1, 6, 50, 64]block_1_past_self_key_states: float32[1, 6, 49, 64]block_1_past_self_value_states: float32[1, 6, 49, 64]block_1_cross_key_states: float32[1, 6, 50, 64]block_1_cross_value_states: float32[1, 6, 50, 64]block_2_past_self_key_states: float32[1, 6, 49, 64]block_2_past_self_value_states: float32[1, 6, 49, 64]block_2_cross_key_states: float32[1, 6, 50, 64]block_2_cross_value_states: float32[1, 6, 50, 64]block_3_past_self_key_states: float32[1, 6, 49, 64]block_3_past_self_value_states: float32[1, 6, 49, 64]block_3_cross_key_states: float32[1, 6, 50, 64]block_3_cross_value_states: float32[1, 6, 50, 64]Completion Time
2/22/2026, 11:03:31 AM
Options
--qairt_version 2.42Versions
- 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
563 μs
Estimated Peak Memory Usage
1 ‑ 4 MB
Compute Units
NPU
197
| Stage | Time | Memory |
|---|---|---|
First App Load | 210 ms | 21‑22 MB |
Subsequent App Load | 203 ms | 12‑13 MB |
Inference | 563 μs | 1‑4 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