Qualcomm® AI HubAI Hub

Profile Job Results

Jobs
jg9zex7mp
Results Ready
Name
whisper_small_decoder
Target Device
  • SA7255P ADP
  • Android 14
  • Qualcomm® SA7255P
Creator
ai-hub-support@qti.qualcomm.com
Input Specs
input_ids: int32[1, 1]
position_ids: int32[1]
k_cache_self_0_in: float16[12, 1, 64, 199]
v_cache_self_0_in: float16[12, 1, 199, 64]
attention_mask: float16[1, 1, 1, 200]
k_cache_cross_0: float16[12, 1, 64, 1500]
v_cache_cross_0: float16[12, 1, 1500, 64]
k_cache_self_1_in: float16[12, 1, 64, 199]
v_cache_self_1_in: float16[12, 1, 199, 64]
k_cache_cross_1: float16[12, 1, 64, 1500]
v_cache_cross_1: float16[12, 1, 1500, 64]
k_cache_self_2_in: float16[12, 1, 64, 199]
v_cache_self_2_in: float16[12, 1, 199, 64]
k_cache_cross_2: float16[12, 1, 64, 1500]
v_cache_cross_2: float16[12, 1, 1500, 64]
k_cache_self_3_in: float16[12, 1, 64, 199]
v_cache_self_3_in: float16[12, 1, 199, 64]
k_cache_cross_3: float16[12, 1, 64, 1500]
v_cache_cross_3: float16[12, 1, 1500, 64]
k_cache_self_4_in: float16[12, 1, 64, 199]
v_cache_self_4_in: float16[12, 1, 199, 64]
k_cache_cross_4: float16[12, 1, 64, 1500]
v_cache_cross_4: float16[12, 1, 1500, 64]
k_cache_self_5_in: float16[12, 1, 64, 199]
v_cache_self_5_in: float16[12, 1, 199, 64]
k_cache_cross_5: float16[12, 1, 64, 1500]
v_cache_cross_5: float16[12, 1, 1500, 64]
k_cache_self_6_in: float16[12, 1, 64, 199]
v_cache_self_6_in: float16[12, 1, 199, 64]
k_cache_cross_6: float16[12, 1, 64, 1500]
v_cache_cross_6: float16[12, 1, 1500, 64]
k_cache_self_7_in: float16[12, 1, 64, 199]
v_cache_self_7_in: float16[12, 1, 199, 64]
k_cache_cross_7: float16[12, 1, 64, 1500]
v_cache_cross_7: float16[12, 1, 1500, 64]
k_cache_self_8_in: float16[12, 1, 64, 199]
v_cache_self_8_in: float16[12, 1, 199, 64]
k_cache_cross_8: float16[12, 1, 64, 1500]
v_cache_cross_8: float16[12, 1, 1500, 64]
k_cache_self_9_in: float16[12, 1, 64, 199]
v_cache_self_9_in: float16[12, 1, 199, 64]
k_cache_cross_9: float16[12, 1, 64, 1500]
v_cache_cross_9: float16[12, 1, 1500, 64]
k_cache_self_10_in: float16[12, 1, 64, 199]
v_cache_self_10_in: float16[12, 1, 199, 64]
k_cache_cross_10: float16[12, 1, 64, 1500]
v_cache_cross_10: float16[12, 1, 1500, 64]
k_cache_self_11_in: float16[12, 1, 64, 199]
v_cache_self_11_in: float16[12, 1, 199, 64]
k_cache_cross_11: float16[12, 1, 64, 1500]
v_cache_cross_11: float16[12, 1, 1500, 64]
Completion Time
10/3/2026, 9:07:39 AM
Versions
  • QAIRT: v2.50.0.260828221209
  • QNN Backend API: 5.50.0
  • QNN Core API: 2.39.0
  • Android: 14 (UAG2.240908.001)
  • Build ID: Snapdragon_Auto.HQX.4.5.6.0.r2-00016-STD.PROD-3
  • AI Hub: aihub-2026.09.25.0
Estimated Inference Time
19.1 ms
Estimated Peak Memory Usage
49 ‑ 58 MB
Compute Units
NPU
2277
StageTimeMemory
First App Load
2.12 s2‑11 MB
Subsequent App Load
997 ms1‑9 MB
Inference
19.1 ms49‑58 MB
QNNValue
default_graph_options.htp_options.optimizations[0].typeFINALIZE_OPTIMIZATION_FLAG
default_graph_options.htp_options.optimizations[0].value3.0
default_graph_options.htp_options.precisionFLOAT16
default_graph_options.htp_options.vtcm_size0

Sign up to run this model on a hosted Qualcomm® device!

Run on device