Profile Job Results
Jobs
jglwjn6mp
Results Ready
Name
opus_mt_en_ko_decoder
Target Device
- QCS8550 (Proxy)
- Android 12
- Qualcomm® Dragonwing™ QCS8550
Creator
ai-hub-support@qti.qualcomm.com
Input Specs
block_0_cross_key_states: float32[1, 16, 256, 64]block_0_cross_value_states: float32[1, 16, 256, 64]block_1_cross_key_states: float32[1, 16, 256, 64]block_1_cross_value_states: float32[1, 16, 256, 64]block_2_cross_key_states: float32[1, 16, 256, 64]block_2_cross_value_states: float32[1, 16, 256, 64]block_3_cross_key_states: float32[1, 16, 256, 64]block_3_cross_value_states: float32[1, 16, 256, 64]block_4_cross_key_states: float32[1, 16, 256, 64]block_4_cross_value_states: float32[1, 16, 256, 64]block_5_cross_key_states: float32[1, 16, 256, 64]block_5_cross_value_states: float32[1, 16, 256, 64]encoder_attention_mask: int32[1, 256]input_ids: int32[1, 1]position: int32[1]block_0_past_self_key_states: float32[1, 16, 255, 64]block_0_past_self_value_states: float32[1, 16, 255, 64]block_1_past_self_key_states: float32[1, 16, 255, 64]block_1_past_self_value_states: float32[1, 16, 255, 64]block_2_past_self_key_states: float32[1, 16, 255, 64]block_2_past_self_value_states: float32[1, 16, 255, 64]block_3_past_self_key_states: float32[1, 16, 255, 64]block_3_past_self_value_states: float32[1, 16, 255, 64]block_4_past_self_key_states: float32[1, 16, 255, 64]block_4_past_self_value_states: float32[1, 16, 255, 64]block_5_past_self_key_states: float32[1, 16, 255, 64]block_5_past_self_value_states: float32[1, 16, 255, 64]Completion Time
10/20/2026, 6:53:13 AM
Options
--qairt_version 2.50Versions
- QAIRT: v2.50.0.260828221209
- QNN Backend API: 5.50.0
- QNN Core API: 2.39.0
- Android: 14 (UP1A.231005.007)
- AI Hub: aihub-2026.10.10.0
Estimated Inference Time
4.70 ms
Estimated Peak Memory Usage
24 ‑ 25 MB
Compute Units
NPU
390
| Stage | Time | Memory |
|---|---|---|
First App Load | 282 ms | 2‑3 MB |
Subsequent App Load | 247 ms | 1‑3 MB |
Inference | 4.70 ms | 24‑25 MB |
| QNN | Value |
|---|---|
| context_options.htp_options.performance_mode | BURST |
| default_graph_options.htp_options.optimizations[0].type | FINALIZE_OPTIMIZATION_FLAG |
| default_graph_options.htp_options.optimizations[0].value | 3.0 |
| default_graph_options.htp_options.precision | FLOAT16 |
| default_graph_options.htp_options.vtcm_size | 0 |
Sign up to run this model on a hosted Qualcomm® device!
Run on device