Profile Job Results
Jobs
jgdvo9d6g
Results Ready
Name
whisper_small_quantized_w8a16_WhisperSmallEncoderQuantizable
Target Device
- SA8295P ADP
- Android 14
- Qualcomm® SA8295P
Creator
ai-hub-support@qti.qualcomm.com
Target Model
Input Specs
input_features: uint16[1, 80, 3000]layers_0_self_attn_k_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]layers_0_self_attn_v_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]_layers_0_self_attn_v_proj_sha_11_Conv_Conv_bias_dq: float32[64]layers_0_self_attn_k_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]layers_0_self_attn_v_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]_layers_0_self_attn_v_proj_sha_10_Conv_Conv_bias_dq: float32[64]layers_0_self_attn_k_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]layers_0_self_attn_v_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]_layers_0_self_attn_v_proj_sha_9_Conv_Conv_bias_dq: float32[64]layers_0_self_attn_k_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]layers_0_self_attn_v_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]_layers_0_self_attn_v_proj_sha_8_Conv_Conv_bias_dq: float32[64]layers_0_self_attn_k_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]layers_0_self_attn_v_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]_layers_0_self_attn_v_proj_sha_7_Conv_Conv_bias_dq: float32[64]layers_0_self_attn_k_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]layers_0_self_attn_v_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]_layers_0_self_attn_v_proj_sha_6_Conv_Conv_bias_dq: float32[64]layers_0_self_attn_k_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]layers_0_self_attn_v_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]_layers_0_self_attn_v_proj_sha_5_Conv_Conv_bias_dq: float32[64]layers_0_self_attn_k_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]layers_0_self_attn_v_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]_layers_0_self_attn_v_proj_sha_4_Conv_Conv_bias_dq: float32[64]layers_0_self_attn_k_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]layers_0_self_attn_v_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]_layers_0_self_attn_v_proj_sha_3_Conv_Conv_bias_dq: float32[64]layers_0_self_attn_k_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]layers_0_self_attn_v_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]_layers_0_self_attn_v_proj_sha_2_Conv_Conv_bias_dq: float32[64]layers_0_self_attn_k_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]layers_0_self_attn_v_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]_layers_0_self_attn_v_proj_sha_1_Conv_Conv_bias_dq: float32[64]layers_0_self_attn_k_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]layers_0_self_attn_v_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]_layers_0_self_attn_v_proj_sha_0_Conv_Conv_bias_dq: float32[64]layers_1_self_attn_k_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]layers_1_self_attn_v_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]_layers_1_self_attn_v_proj_sha_11_Conv_Conv_bias_dq: float32[64]layers_1_self_attn_k_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]layers_1_self_attn_v_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]_layers_1_self_attn_v_proj_sha_10_Conv_Conv_bias_dq: float32[64]layers_1_self_attn_k_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]layers_1_self_attn_v_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]_layers_1_self_attn_v_proj_sha_9_Conv_Conv_bias_dq: float32[64]layers_1_self_attn_k_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]layers_1_self_attn_v_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]_layers_1_self_attn_v_proj_sha_8_Conv_Conv_bias_dq: float32[64]layers_1_self_attn_k_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]layers_1_self_attn_v_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]_layers_1_self_attn_v_proj_sha_7_Conv_Conv_bias_dq: float32[64]layers_1_self_attn_k_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]layers_1_self_attn_v_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]_layers_1_self_attn_v_proj_sha_6_Conv_Conv_bias_dq: float32[64]layers_1_self_attn_k_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]layers_1_self_attn_v_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]_layers_1_self_attn_v_proj_sha_5_Conv_Conv_bias_dq: float32[64]layers_1_self_attn_k_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]layers_1_self_attn_v_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]_layers_1_self_attn_v_proj_sha_4_Conv_Conv_bias_dq: float32[64]layers_1_self_attn_k_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]layers_1_self_attn_v_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]_layers_1_self_attn_v_proj_sha_3_Conv_Conv_bias_dq: float32[64]layers_1_self_attn_k_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]layers_1_self_attn_v_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]_layers_1_self_attn_v_proj_sha_2_Conv_Conv_bias_dq: float32[64]layers_1_self_attn_k_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]layers_1_self_attn_v_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]_layers_1_self_attn_v_proj_sha_1_Conv_Conv_bias_dq: float32[64]layers_1_self_attn_k_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]layers_1_self_attn_v_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]_layers_1_self_attn_v_proj_sha_0_Conv_Conv_bias_dq: float32[64]layers_2_self_attn_k_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]layers_2_self_attn_v_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]_layers_2_self_attn_v_proj_sha_11_Conv_Conv_bias_dq: float32[64]layers_2_self_attn_k_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]layers_2_self_attn_v_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]_layers_2_self_attn_v_proj_sha_10_Conv_Conv_bias_dq: float32[64]layers_2_self_attn_k_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]layers_2_self_attn_v_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]_layers_2_self_attn_v_proj_sha_9_Conv_Conv_bias_dq: float32[64]layers_2_self_attn_k_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]layers_2_self_attn_v_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]_layers_2_self_attn_v_proj_sha_8_Conv_Conv_bias_dq: float32[64]layers_2_self_attn_k_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]layers_2_self_attn_v_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]_layers_2_self_attn_v_proj_sha_7_Conv_Conv_bias_dq: float32[64]layers_2_self_attn_k_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]layers_2_self_attn_v_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]_layers_2_self_attn_v_proj_sha_6_Conv_Conv_bias_dq: float32[64]layers_2_self_attn_k_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]layers_2_self_attn_v_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]_layers_2_self_attn_v_proj_sha_5_Conv_Conv_bias_dq: float32[64]layers_2_self_attn_k_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]layers_2_self_attn_v_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]_layers_2_self_attn_v_proj_sha_4_Conv_Conv_bias_dq: float32[64]layers_2_self_attn_k_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]layers_2_self_attn_v_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]_layers_2_self_attn_v_proj_sha_3_Conv_Conv_bias_dq: float32[64]layers_2_self_attn_k_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]layers_2_self_attn_v_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]_layers_2_self_attn_v_proj_sha_2_Conv_Conv_bias_dq: float32[64]layers_2_self_attn_k_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]layers_2_self_attn_v_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]_layers_2_self_attn_v_proj_sha_1_Conv_Conv_bias_dq: float32[64]layers_2_self_attn_k_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]layers_2_self_attn_v_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]_layers_2_self_attn_v_proj_sha_0_Conv_Conv_bias_dq: float32[64]layers_3_self_attn_k_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]layers_3_self_attn_v_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]_layers_3_self_attn_v_proj_sha_11_Conv_Conv_bias_dq: float32[64]layers_3_self_attn_k_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]layers_3_self_attn_v_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]_layers_3_self_attn_v_proj_sha_10_Conv_Conv_bias_dq: float32[64]layers_3_self_attn_k_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]layers_3_self_attn_v_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]_layers_3_self_attn_v_proj_sha_9_Conv_Conv_bias_dq: float32[64]layers_3_self_attn_k_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]layers_3_self_attn_v_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]_layers_3_self_attn_v_proj_sha_8_Conv_Conv_bias_dq: float32[64]layers_3_self_attn_k_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]layers_3_self_attn_v_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]_layers_3_self_attn_v_proj_sha_7_Conv_Conv_bias_dq: float32[64]layers_3_self_attn_k_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]layers_3_self_attn_v_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]_layers_3_self_attn_v_proj_sha_6_Conv_Conv_bias_dq: float32[64]layers_3_self_attn_k_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]layers_3_self_attn_v_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]_layers_3_self_attn_v_proj_sha_5_Conv_Conv_bias_dq: float32[64]layers_3_self_attn_k_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]layers_3_self_attn_v_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]_layers_3_self_attn_v_proj_sha_4_Conv_Conv_bias_dq: float32[64]layers_3_self_attn_k_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]layers_3_self_attn_v_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]_layers_3_self_attn_v_proj_sha_3_Conv_Conv_bias_dq: float32[64]layers_3_self_attn_k_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]layers_3_self_attn_v_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]_layers_3_self_attn_v_proj_sha_2_Conv_Conv_bias_dq: float32[64]layers_3_self_attn_k_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]layers_3_self_attn_v_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]_layers_3_self_attn_v_proj_sha_1_Conv_Conv_bias_dq: float32[64]layers_3_self_attn_k_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]layers_3_self_attn_v_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]_layers_3_self_attn_v_proj_sha_0_Conv_Conv_bias_dq: float32[64]layers_4_self_attn_k_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]layers_4_self_attn_v_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]_layers_4_self_attn_v_proj_sha_11_Conv_Conv_bias_dq: float32[64]layers_4_self_attn_k_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]layers_4_self_attn_v_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]_layers_4_self_attn_v_proj_sha_10_Conv_Conv_bias_dq: float32[64]layers_4_self_attn_k_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]layers_4_self_attn_v_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]_layers_4_self_attn_v_proj_sha_9_Conv_Conv_bias_dq: float32[64]layers_4_self_attn_k_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]layers_4_self_attn_v_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]_layers_4_self_attn_v_proj_sha_8_Conv_Conv_bias_dq: float32[64]layers_4_self_attn_k_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]layers_4_self_attn_v_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]_layers_4_self_attn_v_proj_sha_7_Conv_Conv_bias_dq: float32[64]layers_4_self_attn_k_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]layers_4_self_attn_v_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]_layers_4_self_attn_v_proj_sha_6_Conv_Conv_bias_dq: float32[64]layers_4_self_attn_k_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]layers_4_self_attn_v_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]_layers_4_self_attn_v_proj_sha_5_Conv_Conv_bias_dq: float32[64]layers_4_self_attn_k_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]layers_4_self_attn_v_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]_layers_4_self_attn_v_proj_sha_4_Conv_Conv_bias_dq: float32[64]layers_4_self_attn_k_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]layers_4_self_attn_v_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]_layers_4_self_attn_v_proj_sha_3_Conv_Conv_bias_dq: float32[64]layers_4_self_attn_k_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]layers_4_self_attn_v_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]_layers_4_self_attn_v_proj_sha_2_Conv_Conv_bias_dq: float32[64]layers_4_self_attn_k_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]layers_4_self_attn_v_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]_layers_4_self_attn_v_proj_sha_1_Conv_Conv_bias_dq: float32[64]layers_4_self_attn_k_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]layers_4_self_attn_v_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]_layers_4_self_attn_v_proj_sha_0_Conv_Conv_bias_dq: float32[64]layers_5_self_attn_k_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]layers_5_self_attn_v_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]_layers_5_self_attn_v_proj_sha_11_Conv_Conv_bias_dq: float32[64]layers_5_self_attn_k_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]layers_5_self_attn_v_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]_layers_5_self_attn_v_proj_sha_10_Conv_Conv_bias_dq: float32[64]layers_5_self_attn_k_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]layers_5_self_attn_v_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]_layers_5_self_attn_v_proj_sha_9_Conv_Conv_bias_dq: float32[64]layers_5_self_attn_k_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]layers_5_self_attn_v_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]_layers_5_self_attn_v_proj_sha_8_Conv_Conv_bias_dq: float32[64]layers_5_self_attn_k_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]layers_5_self_attn_v_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]_layers_5_self_attn_v_proj_sha_7_Conv_Conv_bias_dq: float32[64]layers_5_self_attn_k_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]layers_5_self_attn_v_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]_layers_5_self_attn_v_proj_sha_6_Conv_Conv_bias_dq: float32[64]layers_5_self_attn_k_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]layers_5_self_attn_v_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]_layers_5_self_attn_v_proj_sha_5_Conv_Conv_bias_dq: float32[64]layers_5_self_attn_k_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]layers_5_self_attn_v_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]_layers_5_self_attn_v_proj_sha_4_Conv_Conv_bias_dq: float32[64]layers_5_self_attn_k_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]layers_5_self_attn_v_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]_layers_5_self_attn_v_proj_sha_3_Conv_Conv_bias_dq: float32[64]layers_5_self_attn_k_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]layers_5_self_attn_v_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]_layers_5_self_attn_v_proj_sha_2_Conv_Conv_bias_dq: float32[64]layers_5_self_attn_k_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]layers_5_self_attn_v_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]_layers_5_self_attn_v_proj_sha_1_Conv_Conv_bias_dq: float32[64]layers_5_self_attn_k_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]layers_5_self_attn_v_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]_layers_5_self_attn_v_proj_sha_0_Conv_Conv_bias_dq: float32[64]layers_6_self_attn_k_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]layers_6_self_attn_v_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]_layers_6_self_attn_v_proj_sha_11_Conv_Conv_bias_dq: float32[64]layers_6_self_attn_k_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]layers_6_self_attn_v_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]_layers_6_self_attn_v_proj_sha_10_Conv_Conv_bias_dq: float32[64]layers_6_self_attn_k_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]layers_6_self_attn_v_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]_layers_6_self_attn_v_proj_sha_9_Conv_Conv_bias_dq: float32[64]layers_6_self_attn_k_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]layers_6_self_attn_v_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]_layers_6_self_attn_v_proj_sha_8_Conv_Conv_bias_dq: float32[64]layers_6_self_attn_k_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]layers_6_self_attn_v_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]_layers_6_self_attn_v_proj_sha_7_Conv_Conv_bias_dq: float32[64]layers_6_self_attn_k_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]layers_6_self_attn_v_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]_layers_6_self_attn_v_proj_sha_6_Conv_Conv_bias_dq: float32[64]layers_6_self_attn_k_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]layers_6_self_attn_v_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]_layers_6_self_attn_v_proj_sha_5_Conv_Conv_bias_dq: float32[64]layers_6_self_attn_k_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]layers_6_self_attn_v_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]_layers_6_self_attn_v_proj_sha_4_Conv_Conv_bias_dq: float32[64]layers_6_self_attn_k_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]layers_6_self_attn_v_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]_layers_6_self_attn_v_proj_sha_3_Conv_Conv_bias_dq: float32[64]layers_6_self_attn_k_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]layers_6_self_attn_v_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]_layers_6_self_attn_v_proj_sha_2_Conv_Conv_bias_dq: float32[64]layers_6_self_attn_k_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]layers_6_self_attn_v_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]_layers_6_self_attn_v_proj_sha_1_Conv_Conv_bias_dq: float32[64]layers_6_self_attn_k_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]layers_6_self_attn_v_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]_layers_6_self_attn_v_proj_sha_0_Conv_Conv_bias_dq: float32[64]layers_7_self_attn_k_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]layers_7_self_attn_v_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]_layers_7_self_attn_v_proj_sha_11_Conv_Conv_bias_dq: float32[64]layers_7_self_attn_k_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]layers_7_self_attn_v_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]_layers_7_self_attn_v_proj_sha_10_Conv_Conv_bias_dq: float32[64]layers_7_self_attn_k_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]layers_7_self_attn_v_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]_layers_7_self_attn_v_proj_sha_9_Conv_Conv_bias_dq: float32[64]layers_7_self_attn_k_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]layers_7_self_attn_v_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]_layers_7_self_attn_v_proj_sha_8_Conv_Conv_bias_dq: float32[64]layers_7_self_attn_k_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]layers_7_self_attn_v_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]_layers_7_self_attn_v_proj_sha_7_Conv_Conv_bias_dq: float32[64]layers_7_self_attn_k_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]layers_7_self_attn_v_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]_layers_7_self_attn_v_proj_sha_6_Conv_Conv_bias_dq: float32[64]layers_7_self_attn_k_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]layers_7_self_attn_v_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]_layers_7_self_attn_v_proj_sha_5_Conv_Conv_bias_dq: float32[64]layers_7_self_attn_k_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]layers_7_self_attn_v_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]_layers_7_self_attn_v_proj_sha_4_Conv_Conv_bias_dq: float32[64]layers_7_self_attn_k_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]layers_7_self_attn_v_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]_layers_7_self_attn_v_proj_sha_3_Conv_Conv_bias_dq: float32[64]layers_7_self_attn_k_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]layers_7_self_attn_v_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]_layers_7_self_attn_v_proj_sha_2_Conv_Conv_bias_dq: float32[64]layers_7_self_attn_k_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]layers_7_self_attn_v_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]_layers_7_self_attn_v_proj_sha_1_Conv_Conv_bias_dq: float32[64]layers_7_self_attn_k_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]layers_7_self_attn_v_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]_layers_7_self_attn_v_proj_sha_0_Conv_Conv_bias_dq: float32[64]layers_8_self_attn_k_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]layers_8_self_attn_v_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]_layers_8_self_attn_v_proj_sha_11_Conv_Conv_bias_dq: float32[64]layers_8_self_attn_k_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]layers_8_self_attn_v_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]_layers_8_self_attn_v_proj_sha_10_Conv_Conv_bias_dq: float32[64]layers_8_self_attn_k_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]layers_8_self_attn_v_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]_layers_8_self_attn_v_proj_sha_9_Conv_Conv_bias_dq: float32[64]layers_8_self_attn_k_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]layers_8_self_attn_v_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]_layers_8_self_attn_v_proj_sha_8_Conv_Conv_bias_dq: float32[64]layers_8_self_attn_k_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]layers_8_self_attn_v_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]_layers_8_self_attn_v_proj_sha_7_Conv_Conv_bias_dq: float32[64]layers_8_self_attn_k_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]layers_8_self_attn_v_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]_layers_8_self_attn_v_proj_sha_6_Conv_Conv_bias_dq: float32[64]layers_8_self_attn_k_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]layers_8_self_attn_v_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]_layers_8_self_attn_v_proj_sha_5_Conv_Conv_bias_dq: float32[64]layers_8_self_attn_k_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]layers_8_self_attn_v_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]_layers_8_self_attn_v_proj_sha_4_Conv_Conv_bias_dq: float32[64]layers_8_self_attn_k_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]layers_8_self_attn_v_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]_layers_8_self_attn_v_proj_sha_3_Conv_Conv_bias_dq: float32[64]layers_8_self_attn_k_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]layers_8_self_attn_v_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]_layers_8_self_attn_v_proj_sha_2_Conv_Conv_bias_dq: float32[64]layers_8_self_attn_k_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]layers_8_self_attn_v_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]_layers_8_self_attn_v_proj_sha_1_Conv_Conv_bias_dq: float32[64]layers_8_self_attn_k_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]layers_8_self_attn_v_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]_layers_8_self_attn_v_proj_sha_0_Conv_Conv_bias_dq: float32[64]layers_9_self_attn_k_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]layers_9_self_attn_v_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]_layers_9_self_attn_v_proj_sha_11_Conv_Conv_bias_dq: float32[64]layers_9_self_attn_k_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]layers_9_self_attn_v_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]_layers_9_self_attn_v_proj_sha_10_Conv_Conv_bias_dq: float32[64]layers_9_self_attn_k_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]layers_9_self_attn_v_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]_layers_9_self_attn_v_proj_sha_9_Conv_Conv_bias_dq: float32[64]layers_9_self_attn_k_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]layers_9_self_attn_v_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]_layers_9_self_attn_v_proj_sha_8_Conv_Conv_bias_dq: float32[64]layers_9_self_attn_k_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]layers_9_self_attn_v_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]_layers_9_self_attn_v_proj_sha_7_Conv_Conv_bias_dq: float32[64]layers_9_self_attn_k_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]layers_9_self_attn_v_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]_layers_9_self_attn_v_proj_sha_6_Conv_Conv_bias_dq: float32[64]layers_9_self_attn_k_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]layers_9_self_attn_v_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]_layers_9_self_attn_v_proj_sha_5_Conv_Conv_bias_dq: float32[64]layers_9_self_attn_k_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]layers_9_self_attn_v_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]_layers_9_self_attn_v_proj_sha_4_Conv_Conv_bias_dq: float32[64]layers_9_self_attn_k_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]layers_9_self_attn_v_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]_layers_9_self_attn_v_proj_sha_3_Conv_Conv_bias_dq: float32[64]layers_9_self_attn_k_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]layers_9_self_attn_v_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]_layers_9_self_attn_v_proj_sha_2_Conv_Conv_bias_dq: float32[64]layers_9_self_attn_k_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]layers_9_self_attn_v_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]_layers_9_self_attn_v_proj_sha_1_Conv_Conv_bias_dq: float32[64]layers_9_self_attn_k_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]layers_9_self_attn_v_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]_layers_9_self_attn_v_proj_sha_0_Conv_Conv_bias_dq: float32[64]layers_10_self_attn_k_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]layers_10_self_attn_v_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]_layers_10_self_attn_v_proj_sha_11_Conv_Conv_bias_dq: float32[64]layers_10_self_attn_k_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]layers_10_self_attn_v_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]_layers_10_self_attn_v_proj_sha_10_Conv_Conv_bias_dq: float32[64]layers_10_self_attn_k_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]layers_10_self_attn_v_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]_layers_10_self_attn_v_proj_sha_9_Conv_Conv_bias_dq: float32[64]layers_10_self_attn_k_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]layers_10_self_attn_v_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]_layers_10_self_attn_v_proj_sha_8_Conv_Conv_bias_dq: float32[64]layers_10_self_attn_k_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]layers_10_self_attn_v_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]_layers_10_self_attn_v_proj_sha_7_Conv_Conv_bias_dq: float32[64]layers_10_self_attn_k_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]layers_10_self_attn_v_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]_layers_10_self_attn_v_proj_sha_6_Conv_Conv_bias_dq: float32[64]layers_10_self_attn_k_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]layers_10_self_attn_v_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]_layers_10_self_attn_v_proj_sha_5_Conv_Conv_bias_dq: float32[64]layers_10_self_attn_k_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]layers_10_self_attn_v_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]_layers_10_self_attn_v_proj_sha_4_Conv_Conv_bias_dq: float32[64]layers_10_self_attn_k_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]layers_10_self_attn_v_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]_layers_10_self_attn_v_proj_sha_3_Conv_Conv_bias_dq: float32[64]layers_10_self_attn_k_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]layers_10_self_attn_v_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]_layers_10_self_attn_v_proj_sha_2_Conv_Conv_bias_dq: float32[64]layers_10_self_attn_k_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]layers_10_self_attn_v_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]_layers_10_self_attn_v_proj_sha_1_Conv_Conv_bias_dq: float32[64]layers_10_self_attn_k_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]layers_10_self_attn_v_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]_layers_10_self_attn_v_proj_sha_0_Conv_Conv_bias_dq: float32[64]layers_11_self_attn_k_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]layers_11_self_attn_v_proj_sha_11_Conv_weight_dq: float32[64, 768, 1, 1]_layers_11_self_attn_v_proj_sha_11_Conv_Conv_bias_dq: float32[64]layers_11_self_attn_k_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]layers_11_self_attn_v_proj_sha_10_Conv_weight_dq: float32[64, 768, 1, 1]_layers_11_self_attn_v_proj_sha_10_Conv_Conv_bias_dq: float32[64]layers_11_self_attn_k_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]layers_11_self_attn_v_proj_sha_9_Conv_weight_dq: float32[64, 768, 1, 1]_layers_11_self_attn_v_proj_sha_9_Conv_Conv_bias_dq: float32[64]layers_11_self_attn_k_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]layers_11_self_attn_v_proj_sha_8_Conv_weight_dq: float32[64, 768, 1, 1]_layers_11_self_attn_v_proj_sha_8_Conv_Conv_bias_dq: float32[64]layers_11_self_attn_k_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]layers_11_self_attn_v_proj_sha_7_Conv_weight_dq: float32[64, 768, 1, 1]_layers_11_self_attn_v_proj_sha_7_Conv_Conv_bias_dq: float32[64]layers_11_self_attn_k_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]layers_11_self_attn_v_proj_sha_6_Conv_weight_dq: float32[64, 768, 1, 1]_layers_11_self_attn_v_proj_sha_6_Conv_Conv_bias_dq: float32[64]layers_11_self_attn_k_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]layers_11_self_attn_v_proj_sha_5_Conv_weight_dq: float32[64, 768, 1, 1]_layers_11_self_attn_v_proj_sha_5_Conv_Conv_bias_dq: float32[64]layers_11_self_attn_k_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]layers_11_self_attn_v_proj_sha_4_Conv_weight_dq: float32[64, 768, 1, 1]_layers_11_self_attn_v_proj_sha_4_Conv_Conv_bias_dq: float32[64]layers_11_self_attn_k_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]layers_11_self_attn_v_proj_sha_3_Conv_weight_dq: float32[64, 768, 1, 1]_layers_11_self_attn_v_proj_sha_3_Conv_Conv_bias_dq: float32[64]layers_11_self_attn_k_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]layers_11_self_attn_v_proj_sha_2_Conv_weight_dq: float32[64, 768, 1, 1]_layers_11_self_attn_v_proj_sha_2_Conv_Conv_bias_dq: float32[64]layers_11_self_attn_k_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]layers_11_self_attn_v_proj_sha_1_Conv_weight_dq: float32[64, 768, 1, 1]_layers_11_self_attn_v_proj_sha_1_Conv_Conv_bias_dq: float32[64]layers_11_self_attn_k_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]layers_11_self_attn_v_proj_sha_0_Conv_weight_dq: float32[64, 768, 1, 1]_layers_11_self_attn_v_proj_sha_0_Conv_Conv_bias_dq: float32[64]Completion Time
1/31/2026, 6:03:21 PM
Options
--max_profiler_iterations 10Versions
- QAIRT: v2.42.0.251225135753_193295
- QNN Backend API: 5.41.0
- QNN Core API: 2.32.0
- Android: 14 (UQ1A.240205.002)
- Build ID: SA8295P.HQX.4.5.6.0-00006-STD.PROD-1
- AI Hub: aihub-2026.01.22.0
Estimated Inference Time
1.36 s
Estimated Peak Memory Usage
26 ‑ 38 MB
Compute Units
NPU
3454
| Stage | Time | Memory |
|---|---|---|
First App Load | 2.52 s | 5‑10 MB |
Subsequent App Load | 2.51 s | 1‑8 MB |
Inference | 1.36 s | 26‑38 MB |
| QNN | Value |
|---|---|
| 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







