Profile Job Results
Jobs
j5m4q1d75
Results Ready
Name
pi05_action_expert
Target Device
- Samsung Galaxy S24
- Android 14
- Snapdragon® 8 Gen 3 | SM8650
Creator
ai-hub-support@qti.qualcomm.com
Input Specs
x_t: float32[1, 50, 32]time_step: float32[1]key_cache_l0: float32[1, 968, 1, 256]key_cache_l1: float32[1, 968, 1, 256]key_cache_l10: float32[1, 968, 1, 256]key_cache_l11: float32[1, 968, 1, 256]key_cache_l12: float32[1, 968, 1, 256]key_cache_l13: float32[1, 968, 1, 256]key_cache_l14: float32[1, 968, 1, 256]key_cache_l15: float32[1, 968, 1, 256]key_cache_l16: float32[1, 968, 1, 256]key_cache_l17: float32[1, 968, 1, 256]key_cache_l2: float32[1, 968, 1, 256]key_cache_l3: float32[1, 968, 1, 256]key_cache_l4: float32[1, 968, 1, 256]key_cache_l5: float32[1, 968, 1, 256]key_cache_l6: float32[1, 968, 1, 256]key_cache_l7: float32[1, 968, 1, 256]key_cache_l8: float32[1, 968, 1, 256]key_cache_l9: float32[1, 968, 1, 256]value_cache_l0: float32[1, 968, 1, 256]value_cache_l1: float32[1, 968, 1, 256]value_cache_l10: float32[1, 968, 1, 256]value_cache_l11: float32[1, 968, 1, 256]value_cache_l12: float32[1, 968, 1, 256]value_cache_l13: float32[1, 968, 1, 256]value_cache_l14: float32[1, 968, 1, 256]value_cache_l15: float32[1, 968, 1, 256]value_cache_l16: float32[1, 968, 1, 256]value_cache_l17: float32[1, 968, 1, 256]value_cache_l2: float32[1, 968, 1, 256]value_cache_l3: float32[1, 968, 1, 256]value_cache_l4: float32[1, 968, 1, 256]value_cache_l5: float32[1, 968, 1, 256]value_cache_l6: float32[1, 968, 1, 256]value_cache_l7: float32[1, 968, 1, 256]value_cache_l8: float32[1, 968, 1, 256]value_cache_l9: float32[1, 968, 1, 256]rope_emb_cos: float32[1, 50, 1, 128]rope_emb_sin: float32[1, 50, 1, 128]full_att_4d: float32[1, 1, 50, 1018]Completion Time
5/30/2026, 7:52:18 AM
Versions
- QAIRT: v2.45.0.260326154327
- QNN Backend API: 5.45.0
- QNN Core API: 2.34.0
- Android: 14 (UP1A.231005.007)
- AI Hub: aihub-2026.05.27.0
Estimated Inference Time
21.7 ms
Estimated Peak Memory Usage
35 ‑ 42 MB
Compute Units
NPU
3835
| Stage | Time | Memory |
|---|---|---|
First App Load | 555 ms | 4‑11 MB |
Subsequent App Load | 346 ms | 1‑9 MB |
Inference | 21.7 ms | 35‑42 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