Profile Job Results
Jobs
jp18jd1ng
Results Ready
Name
opus_mt_zh_en_OpusMTDecoder
Target Device
- QCS8450 (Proxy)
- Android 13
- Qualcomm® QCS8450
Creator
ai-hub-support@qti.qualcomm.com
Target Model
Input Specs
input_ids: int32[1, 1]position: int32[1]encoder_attention_mask: int32[1, 256]block_0_past_self_key_states: float16[1, 8, 255, 64]block_0_past_self_value_states: float16[1, 8, 255, 64]block_0_cross_key_states: float16[1, 8, 256, 64]block_0_cross_value_states: float16[1, 8, 256, 64]block_1_past_self_key_states: float16[1, 8, 255, 64]block_1_past_self_value_states: float16[1, 8, 255, 64]block_1_cross_key_states: float16[1, 8, 256, 64]block_1_cross_value_states: float16[1, 8, 256, 64]block_2_past_self_key_states: float16[1, 8, 255, 64]block_2_past_self_value_states: float16[1, 8, 255, 64]block_2_cross_key_states: float16[1, 8, 256, 64]block_2_cross_value_states: float16[1, 8, 256, 64]block_3_past_self_key_states: float16[1, 8, 255, 64]block_3_past_self_value_states: float16[1, 8, 255, 64]block_3_cross_key_states: float16[1, 8, 256, 64]block_3_cross_value_states: float16[1, 8, 256, 64]block_4_past_self_key_states: float16[1, 8, 255, 64]block_4_past_self_value_states: float16[1, 8, 255, 64]block_4_cross_key_states: float16[1, 8, 256, 64]block_4_cross_value_states: float16[1, 8, 256, 64]block_5_past_self_key_states: float16[1, 8, 255, 64]block_5_past_self_value_states: float16[1, 8, 255, 64]block_5_cross_key_states: float16[1, 8, 256, 64]block_5_cross_value_states: float16[1, 8, 256, 64]Completion Time
2/22/2026, 6:20:58 PM
Versions
- QAIRT: v2.43.0.260127150333_193827
- QNN Backend API: 5.43.0
- QNN Core API: 2.32.0
- Android: 13 (TP1A.220624.014)
- AI Hub: aihub-2026.02.14.0
Estimated Inference Time
4.00 ms
Estimated Peak Memory Usage
6 ‑ 173 MB
Compute Units
NPU
349
| Stage | Time | Memory |
|---|---|---|
First App Load | 3.80 s | 606‑614 MB |
Subsequent App Load | 240 ms | 1‑170 MB |
Inference | 4.00 ms | 6‑173 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