Qualcomm® AI HubAI Hub

Profile Job Results

Jobs
jp2zjjomp
Results Ready
Name
whisper_tiny_decoder
Target Device
  • Dragonwing IQ-9075 EVK
  • Qualcomm Linux 1.9
  • Qualcomm® Dragonwing™ IQ-9075 | QCS9075
Creator
ai-hub-support@qti.qualcomm.com
Input Specs
input_ids: int32[1, 1]
position_ids: int32[1]
k_cache_self_0_in: float16[6, 1, 64, 199]
attention_mask: float16[1, 1, 1, 200]
v_cache_self_0_in: float16[6, 1, 199, 64]
k_cache_cross_0: float16[6, 1, 64, 1500]
v_cache_cross_0: float16[6, 1, 1500, 64]
k_cache_self_1_in: float16[6, 1, 64, 199]
v_cache_self_1_in: float16[6, 1, 199, 64]
k_cache_cross_1: float16[6, 1, 64, 1500]
v_cache_cross_1: float16[6, 1, 1500, 64]
k_cache_self_2_in: float16[6, 1, 64, 199]
v_cache_self_2_in: float16[6, 1, 199, 64]
k_cache_cross_2: float16[6, 1, 64, 1500]
v_cache_cross_2: float16[6, 1, 1500, 64]
k_cache_self_3_in: float16[6, 1, 64, 199]
v_cache_self_3_in: float16[6, 1, 199, 64]
k_cache_cross_3: float16[6, 1, 64, 1500]
v_cache_cross_3: float16[6, 1, 1500, 64]
Completion Time
11/14/2026, 4:06:34 PM
Options
--qairt_version latest
Versions
  • ONNX Runtime: 1.30.0
  • QNN Execution Provider: 2.5.0
  • QAIRT: v2.51.0.260929181845
  • Qualcomm Linux: 1.9-ver.1.0
  • AI Hub: aihub-2026.11.06.1
Estimated Inference Time
2.95 ms
Estimated Peak Memory Usage
10 ‑ 23 MB
Compute Units
NPU
531
StageTimeMemory
First App Load
307 ms107‑108 MB
Subsequent App Load
290 ms109‑110 MB
Inference
2.95 ms10‑23 MB
ONNX RuntimeValue
execution_modeSEQUENTIAL
intra_op_num_threads0
inter_op_num_threads0
enable_memory_patternfalse
enable_cpu_memory_arenafalse
graph_optimization_levelENABLE_ALL
QNN Execution ProviderValue
htp_performance_mode"burst"
htp_graph_finalization_optimization_mode"3"
enable_htp_fp16_precision"1"
capture_network_visualizationsfalse
context_priority"normal"
offload_graph_io_quantization"1"
num_graph_prepare_threads"1"

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

Run on device