Qualcomm® AI HubAI Hub

Profile Job Results

Jobs
jg9ovrymg
Results Ready
Name
melotts_ja_charsiu_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
encoder_attention_mask: int32[1, 50]
block_0_cross_key_states: uint16[1, 6, 50, 64]
block_1_cross_key_states: uint16[1, 6, 50, 64]
block_2_cross_key_states: uint16[1, 6, 50, 64]
block_3_cross_key_states: uint16[1, 6, 50, 64]
input_ids: int32[1, 1]
position: int32[1, 1]
block_0_past_self_key_states: uint16[1, 6, 49, 64]
block_0_past_self_value_states: uint16[1, 6, 49, 64]
block_0_cross_value_states: uint16[1, 6, 50, 64]
block_1_past_self_key_states: uint16[1, 6, 49, 64]
block_1_past_self_value_states: uint16[1, 6, 49, 64]
block_1_cross_value_states: uint16[1, 6, 50, 64]
block_2_past_self_key_states: uint16[1, 6, 49, 64]
block_2_past_self_value_states: uint16[1, 6, 49, 64]
block_2_cross_value_states: uint16[1, 6, 50, 64]
block_3_past_self_key_states: uint16[1, 6, 49, 64]
block_3_past_self_value_states: uint16[1, 6, 49, 64]
block_3_cross_value_states: uint16[1, 6, 50, 64]
Completion Time
10/20/2026, 9:49:25 PM
Options
--qairt_version 2.50
Versions
  • QAIRT: v2.50.0.260828221209
  • QNN Backend API: 5.50.0
  • QNN Core API: 2.39.0
  • Qualcomm Linux: 1.9-ver.1.0
  • AI Hub: aihub-2026.10.10.0
Estimated Inference Time
604 μs
Estimated Peak Memory Usage
0 ‑ 2 MB
Compute Units
NPU
200
StageTimeMemory
First App Load
207 ms4‑5 MB
Subsequent App Load
155 ms2‑3 MB
Inference
604 μs0‑2 MB
QNNValue
context_options.htp_options.performance_modeBURST
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