Other
PyTorch
android
qaihm-bot commited on
Commit
14f47de
·
verified ·
1 Parent(s): 5915705

See https://github.com/qualcomm/ai-hub-models/releases/v0.56.0 for changelog.

Files changed (2) hide show
  1. README.md +44 -59
  2. release_assets.json +41 -101
README.md CHANGED
@@ -27,31 +27,25 @@ Below are pre-exported model assets ready for deployment.
27
 
28
  | Runtime | Precision | Chipset | SDK Versions | Download |
29
  |---|---|---|---|---|
30
- | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | QAIRT 2.42, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-float-qualcomm_snapdragon_8_elite_gen5.zip)
31
- | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | QAIRT 2.42, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-float-qualcomm_snapdragon_x2_elite.zip)
32
- | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | QAIRT 2.42, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-float-qualcomm_snapdragon_x_elite.zip)
33
- | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.42, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-float-qualcomm_snapdragon_8gen3.zip)
34
- | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS8550 (Proxy) | QAIRT 2.42, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-float-qualcomm_qcs8550_proxy.zip)
35
- | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | QAIRT 2.42, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-float-qualcomm_snapdragon_8_elite_for_galaxy.zip)
36
- | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS9075 | QAIRT 2.42, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-float-qualcomm_qcs9075.zip)
37
- | PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Elite Gen 5 Mobile | QAIRT 2.42, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-w8a16_mixed_fp16-qualcomm_snapdragon_8_elite_gen5.zip)
38
- | PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Snapdragon® X2 Elite | QAIRT 2.42, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-w8a16_mixed_fp16-qualcomm_snapdragon_x2_elite.zip)
39
- | PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Snapdragon® X Elite | QAIRT 2.42, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-w8a16_mixed_fp16-qualcomm_snapdragon_x_elite.zip)
40
- | PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.42, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-w8a16_mixed_fp16-qualcomm_snapdragon_8gen3.zip)
41
- | PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Qualcomm® QCS8550 (Proxy) | QAIRT 2.42, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-w8a16_mixed_fp16-qualcomm_qcs8550_proxy.zip)
42
- | PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Elite For Galaxy Mobile | QAIRT 2.42, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-w8a16_mixed_fp16-qualcomm_snapdragon_8_elite_for_galaxy.zip)
43
- | PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Qualcomm® QCS9075 | QAIRT 2.42, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-w8a16_mixed_fp16-qualcomm_qcs9075.zip)
44
- | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-qnn_context_binary-float-qualcomm_snapdragon_8_elite_gen5.zip)
45
- | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-qnn_context_binary-float-qualcomm_snapdragon_x2_elite.zip)
46
- | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-qnn_context_binary-float-qualcomm_snapdragon_x_elite.zip)
47
- | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-qnn_context_binary-float-qualcomm_snapdragon_8gen3.zip)
48
- | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8550 (Proxy) | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-qnn_context_binary-float-qualcomm_qcs8550_proxy.zip)
49
- | QNN_CONTEXT_BINARY | float | Qualcomm® SA8775P | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-qnn_context_binary-float-qualcomm_sa8775p.zip)
50
- | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-qnn_context_binary-float-qualcomm_snapdragon_8_elite_for_galaxy.zip)
51
- | QNN_CONTEXT_BINARY | float | Qualcomm® SA7255P | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-qnn_context_binary-float-qualcomm_sa7255p.zip)
52
- | QNN_CONTEXT_BINARY | float | Qualcomm® SA8295P | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-qnn_context_binary-float-qualcomm_sa8295p.zip)
53
- | QNN_CONTEXT_BINARY | float | Qualcomm® QCS9075 | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-qnn_context_binary-float-qualcomm_qcs9075.zip)
54
- | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8450 (Proxy) | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-qnn_context_binary-float-qualcomm_qcs8450_proxy.zip)
55
 
56
  For more device-specific assets and performance metrics, visit **[GKT on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/gkt)**.
57
 
@@ -80,40 +74,31 @@ See our repository for [GKT on GitHub](https://github.com/qualcomm/ai-hub-models
80
  ## Performance Summary
81
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
82
  |---|---|---|---|---|---|---
83
- | GKT | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 56.903 ms | 8 - 18 MB | NPU
84
- | GKT | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | 62.042 ms | 9 - 9 MB | NPU
85
- | GKT | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | 108.068 ms | 7 - 7 MB | NPU
86
- | GKT | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | 79.759 ms | 8 - 15 MB | NPU
87
- | GKT | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS8550 (Proxy) | 111.333 ms | 8 - 12 MB | NPU
88
- | GKT | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS9075 | 110.854 ms | 7 - 10 MB | NPU
89
- | GKT | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 71.062 ms | 1 - 9 MB | NPU
90
- | GKT | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS8750 | 71.062 ms | 1 - 9 MB | NPU
91
- | GKT | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS7181 | 108.068 ms | 7 - 7 MB | NPU
92
- | GKT | PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Elite Gen 5 Mobile | 37.164 ms | 4 - 14 MB | NPU
93
- | GKT | PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Snapdragon® X2 Elite | 39.965 ms | 4 - 4 MB | NPU
94
- | GKT | PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Snapdragon® X Elite | 87.059 ms | 6 - 6 MB | NPU
95
- | GKT | PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Gen 3 Mobile | 59.337 ms | 4 - 11 MB | NPU
96
- | GKT | PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Qualcomm® QCS8550 (Proxy) | 87.897 ms | 0 - 8 MB | NPU
97
- | GKT | PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Snapdragon® 8 Elite For Galaxy Mobile | 45.356 ms | 0 - 8 MB | NPU
98
- | GKT | PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Qualcomm® QCS9075 | 88.293 ms | 4 - 6 MB | NPU
99
- | GKT | PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Qualcomm® QCS8750 | 45.356 ms | 0 - 8 MB | NPU
100
- | GKT | PRECOMPILED_QNN_ONNX | w8a16_mixed_fp16 | Qualcomm® QCS7181 | 87.059 ms | 6 - 6 MB | NPU
101
- | GKT | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | 58.928 ms | 8 - 17 MB | NPU
102
- | GKT | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | 57.465 ms | 7 - 7 MB | NPU
103
- | GKT | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | 106.08 ms | 7 - 7 MB | NPU
104
- | GKT | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | 81.106 ms | 8 - 16 MB | NPU
105
- | GKT | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8275 | 190.598 ms | 1 - 11 MB | NPU
106
- | GKT | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8550 (Proxy) | 110.968 ms | 9 - 10 MB | NPU
107
- | GKT | QNN_CONTEXT_BINARY | float | Qualcomm® SA8775P | 111.467 ms | 0 - 10 MB | NPU
108
- | GKT | QNN_CONTEXT_BINARY | float | Qualcomm® SA8650P | 111.467 ms | 0 - 10 MB | NPU
109
- | GKT | QNN_CONTEXT_BINARY | float | Qualcomm® SA8255P | 111.467 ms | 0 - 10 MB | NPU
110
- | GKT | QNN_CONTEXT_BINARY | float | Qualcomm® QCS9075 | 110.124 ms | 7 - 17 MB | NPU
111
- | GKT | QNN_CONTEXT_BINARY | float | Qualcomm® SA7255P | 190.598 ms | 1 - 11 MB | NPU
112
- | GKT | QNN_CONTEXT_BINARY | float | Qualcomm® SA8295P | 140.633 ms | 0 - 6 MB | NPU
113
- | GKT | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | 70.743 ms | 7 - 20 MB | NPU
114
- | GKT | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8450 (Proxy) | 205.997 ms | 8 - 17 MB | NPU
115
- | GKT | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8750 | 70.743 ms | 7 - 20 MB | NPU
116
- | GKT | QNN_CONTEXT_BINARY | float | Qualcomm® QCS7181 | 106.08 ms | 7 - 7 MB | NPU
117
 
118
  ## License
119
  * The license for the original implementation of GKT can be found
 
27
 
28
  | Runtime | Precision | Chipset | SDK Versions | Download |
29
  |---|---|---|---|---|
30
+ | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | QAIRT 2.45, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-precompiled_qnn_onnx-float-qualcomm_snapdragon_x2_elite.zip)
31
+ | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | QAIRT 2.45, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-precompiled_qnn_onnx-float-qualcomm_snapdragon_x_elite.zip)
32
+ | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.45, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-precompiled_qnn_onnx-float-qualcomm_snapdragon_8gen3.zip)
33
+ | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 1 Mobile | QAIRT 2.45, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-precompiled_qnn_onnx-float-qualcomm_snapdragon_8gen1.zip)
34
+ | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS8550 (Proxy) | QAIRT 2.45, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-precompiled_qnn_onnx-float-qualcomm_qcs8550_proxy.zip)
35
+ | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Mobile | QAIRT 2.45, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-precompiled_qnn_onnx-float-qualcomm_snapdragon_8_elite_for_galaxy.zip)
36
+ | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | QAIRT 2.45, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-precompiled_qnn_onnx-float-qualcomm_snapdragon_8_elite_gen5_for_galaxy.zip)
37
+ | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS9075 | QAIRT 2.45, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-precompiled_qnn_onnx-float-qualcomm_qcs9075.zip)
38
+ | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-qnn_context_binary-float-qualcomm_snapdragon_x2_elite.zip)
39
+ | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-qnn_context_binary-float-qualcomm_snapdragon_x_elite.zip)
40
+ | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-qnn_context_binary-float-qualcomm_snapdragon_8gen3.zip)
41
+ | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 1 Mobile | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-qnn_context_binary-float-qualcomm_snapdragon_8gen1.zip)
42
+ | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8550 (Proxy) | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-qnn_context_binary-float-qualcomm_qcs8550_proxy.zip)
43
+ | QNN_CONTEXT_BINARY | float | Qualcomm® SA8775P | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-qnn_context_binary-float-qualcomm_sa8775p.zip)
44
+ | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Mobile | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-qnn_context_binary-float-qualcomm_snapdragon_8_elite_for_galaxy.zip)
45
+ | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-qnn_context_binary-float-qualcomm_snapdragon_8_elite_gen5_for_galaxy.zip)
46
+ | QNN_CONTEXT_BINARY | float | Qualcomm® SA7255P | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-qnn_context_binary-float-qualcomm_sa7255p.zip)
47
+ | QNN_CONTEXT_BINARY | float | Qualcomm® SA8295P | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-qnn_context_binary-float-qualcomm_sa8295p.zip)
48
+ | QNN_CONTEXT_BINARY | float | Qualcomm® QCS9075 | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-qnn_context_binary-float-qualcomm_qcs9075.zip)
 
 
 
 
 
 
49
 
50
  For more device-specific assets and performance metrics, visit **[GKT on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/gkt)**.
51
 
 
74
  ## Performance Summary
75
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
76
  |---|---|---|---|---|---|---
77
+ | GKT | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | 57.734 ms | 9 - 9 MB | NPU
78
+ | GKT | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | 107.025 ms | 7 - 7 MB | NPU
79
+ | GKT | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | 81.143 ms | 8 - 14 MB | NPU
80
+ | GKT | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 1 Mobile | 198.13 ms | 8 - 22 MB | NPU
81
+ | GKT | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS8550 (Proxy) | 117.021 ms | 8 - 12 MB | NPU
82
+ | GKT | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS8450 | 198.13 ms | 8 - 22 MB | NPU
83
+ | GKT | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 58.48 ms | 1 - 9 MB | NPU
84
+ | GKT | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS9075 | 110.005 ms | 7 - 10 MB | NPU
85
+ | GKT | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Mobile | 71.236 ms | 1 - 8 MB | NPU
86
+ | GKT | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS8750 | 71.236 ms | 1 - 8 MB | NPU
87
+ | GKT | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS7181 | 107.025 ms | 7 - 7 MB | NPU
88
+ | GKT | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | 57.797 ms | 7 - 7 MB | NPU
89
+ | GKT | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | 107.024 ms | 7 - 7 MB | NPU
90
+ | GKT | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | 82.091 ms | 8 - 15 MB | NPU
91
+ | GKT | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 1 Mobile | 202.253 ms | 6 - 16 MB | NPU
92
+ | GKT | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8275 | 189.362 ms | 0 - 10 MB | NPU
93
+ | GKT | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8550 (Proxy) | 116.843 ms | 8 - 9 MB | NPU
94
+ | GKT | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8450 | 202.253 ms | 6 - 16 MB | NPU
95
+ | GKT | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | 58.625 ms | 7 - 16 MB | NPU
96
+ | GKT | QNN_CONTEXT_BINARY | float | Qualcomm® SA7255P | 189.362 ms | 0 - 10 MB | NPU
97
+ | GKT | QNN_CONTEXT_BINARY | float | Qualcomm® QCS9075 | 110.18 ms | 7 - 17 MB | NPU
98
+ | GKT | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Mobile | 71.321 ms | 7 - 20 MB | NPU
99
+ | GKT | QNN_CONTEXT_BINARY | float | Qualcomm® SA8295P | 140.797 ms | 0 - 6 MB | NPU
100
+ | GKT | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8750 | 71.321 ms | 7 - 20 MB | NPU
101
+ | GKT | QNN_CONTEXT_BINARY | float | Qualcomm® QCS7181 | 107.024 ms | 7 - 7 MB | NPU
 
 
 
 
 
 
 
 
 
102
 
103
  ## License
104
  * The license for the original implementation of GKT can be found
release_assets.json CHANGED
@@ -1,21 +1,36 @@
1
  {
2
- "version": "0.55.0",
3
  "precisions": {
4
  "float": {
5
  "chipset_assets": {
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
6
  "qualcomm-snapdragon-8gen3": {
7
  "qnn_context_binary": {
8
  "tool_versions": {
9
  "qairt": "2.45.0.260326154327"
10
  },
11
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-qnn_context_binary-float-qualcomm_snapdragon_8gen3.zip"
12
  },
13
  "precompiled_qnn_onnx": {
14
  "tool_versions": {
15
- "qairt": "2.42.0.251225135753_193295",
16
  "onnx_runtime": "1.25.0"
17
  },
18
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-float-qualcomm_snapdragon_8gen3.zip"
19
  }
20
  },
21
  "qualcomm-snapdragon-8-elite-for-galaxy": {
@@ -23,29 +38,29 @@
23
  "tool_versions": {
24
  "qairt": "2.45.0.260326154327"
25
  },
26
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-qnn_context_binary-float-qualcomm_snapdragon_8_elite_for_galaxy.zip"
27
  },
28
  "precompiled_qnn_onnx": {
29
  "tool_versions": {
30
- "qairt": "2.42.0.251225135753_193295",
31
  "onnx_runtime": "1.25.0"
32
  },
33
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-float-qualcomm_snapdragon_8_elite_for_galaxy.zip"
34
  }
35
  },
36
- "qualcomm-snapdragon-8-elite-gen5": {
37
  "qnn_context_binary": {
38
  "tool_versions": {
39
  "qairt": "2.45.0.260326154327"
40
  },
41
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-qnn_context_binary-float-qualcomm_snapdragon_8_elite_gen5.zip"
42
  },
43
  "precompiled_qnn_onnx": {
44
  "tool_versions": {
45
- "qairt": "2.42.0.251225135753_193295",
46
  "onnx_runtime": "1.25.0"
47
  },
48
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-float-qualcomm_snapdragon_8_elite_gen5.zip"
49
  }
50
  },
51
  "qualcomm-snapdragon-x-elite": {
@@ -53,14 +68,14 @@
53
  "tool_versions": {
54
  "qairt": "2.45.0.260326154327"
55
  },
56
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-qnn_context_binary-float-qualcomm_snapdragon_x_elite.zip"
57
  },
58
  "precompiled_qnn_onnx": {
59
  "tool_versions": {
60
- "qairt": "2.42.0.251225135753_193295",
61
  "onnx_runtime": "1.25.0"
62
  },
63
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-float-qualcomm_snapdragon_x_elite.zip"
64
  }
65
  },
66
  "qualcomm-snapdragon-x2-elite": {
@@ -68,14 +83,14 @@
68
  "tool_versions": {
69
  "qairt": "2.45.0.260326154327"
70
  },
71
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-qnn_context_binary-float-qualcomm_snapdragon_x2_elite.zip"
72
  },
73
  "precompiled_qnn_onnx": {
74
  "tool_versions": {
75
- "qairt": "2.42.0.251225135753_193295",
76
  "onnx_runtime": "1.25.0"
77
  },
78
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-float-qualcomm_snapdragon_x2_elite.zip"
79
  }
80
  },
81
  "qualcomm-sa7255p": {
@@ -83,7 +98,7 @@
83
  "tool_versions": {
84
  "qairt": "2.45.0.260326154327"
85
  },
86
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-qnn_context_binary-float-qualcomm_sa7255p.zip"
87
  }
88
  },
89
  "qualcomm-sa8295p": {
@@ -91,7 +106,7 @@
91
  "tool_versions": {
92
  "qairt": "2.45.0.260326154327"
93
  },
94
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-qnn_context_binary-float-qualcomm_sa8295p.zip"
95
  }
96
  },
97
  "qualcomm-sa8775p": {
@@ -99,7 +114,7 @@
99
  "tool_versions": {
100
  "qairt": "2.45.0.260326154327"
101
  },
102
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-qnn_context_binary-float-qualcomm_sa8775p.zip"
103
  }
104
  },
105
  "qualcomm-qcs8550-proxy": {
@@ -107,14 +122,14 @@
107
  "tool_versions": {
108
  "qairt": "2.45.0.260326154327"
109
  },
110
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-qnn_context_binary-float-qualcomm_qcs8550_proxy.zip"
111
  },
112
  "precompiled_qnn_onnx": {
113
  "tool_versions": {
114
- "qairt": "2.42.0.251225135753_193295",
115
  "onnx_runtime": "1.25.0"
116
  },
117
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-float-qualcomm_qcs8550_proxy.zip"
118
  }
119
  },
120
  "qualcomm-qcs9075": {
@@ -122,89 +137,14 @@
122
  "tool_versions": {
123
  "qairt": "2.45.0.260326154327"
124
  },
125
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-qnn_context_binary-float-qualcomm_qcs9075.zip"
126
  },
127
  "precompiled_qnn_onnx": {
128
  "tool_versions": {
129
- "qairt": "2.42.0.251225135753_193295",
130
- "onnx_runtime": "1.25.0"
131
- },
132
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-float-qualcomm_qcs9075.zip"
133
- }
134
- },
135
- "qualcomm-qcs8450-proxy": {
136
- "qnn_context_binary": {
137
- "tool_versions": {
138
- "qairt": "2.45.0.260326154327"
139
- },
140
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-qnn_context_binary-float-qualcomm_qcs8450_proxy.zip"
141
- }
142
- }
143
- }
144
- },
145
- "w8a16_mixed_fp16": {
146
- "chipset_assets": {
147
- "qualcomm-snapdragon-8gen3": {
148
- "precompiled_qnn_onnx": {
149
- "tool_versions": {
150
- "qairt": "2.42.0.251225135753_193295",
151
- "onnx_runtime": "1.25.0"
152
- },
153
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-w8a16_mixed_fp16-qualcomm_snapdragon_8gen3.zip"
154
- }
155
- },
156
- "qualcomm-snapdragon-8-elite-for-galaxy": {
157
- "precompiled_qnn_onnx": {
158
- "tool_versions": {
159
- "qairt": "2.42.0.251225135753_193295",
160
- "onnx_runtime": "1.25.0"
161
- },
162
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-w8a16_mixed_fp16-qualcomm_snapdragon_8_elite_for_galaxy.zip"
163
- }
164
- },
165
- "qualcomm-snapdragon-8-elite-gen5": {
166
- "precompiled_qnn_onnx": {
167
- "tool_versions": {
168
- "qairt": "2.42.0.251225135753_193295",
169
- "onnx_runtime": "1.25.0"
170
- },
171
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-w8a16_mixed_fp16-qualcomm_snapdragon_8_elite_gen5.zip"
172
- }
173
- },
174
- "qualcomm-snapdragon-x-elite": {
175
- "precompiled_qnn_onnx": {
176
- "tool_versions": {
177
- "qairt": "2.42.0.251225135753_193295",
178
- "onnx_runtime": "1.25.0"
179
- },
180
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-w8a16_mixed_fp16-qualcomm_snapdragon_x_elite.zip"
181
- }
182
- },
183
- "qualcomm-snapdragon-x2-elite": {
184
- "precompiled_qnn_onnx": {
185
- "tool_versions": {
186
- "qairt": "2.42.0.251225135753_193295",
187
- "onnx_runtime": "1.25.0"
188
- },
189
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-w8a16_mixed_fp16-qualcomm_snapdragon_x2_elite.zip"
190
- }
191
- },
192
- "qualcomm-qcs8550-proxy": {
193
- "precompiled_qnn_onnx": {
194
- "tool_versions": {
195
- "qairt": "2.42.0.251225135753_193295",
196
- "onnx_runtime": "1.25.0"
197
- },
198
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-w8a16_mixed_fp16-qualcomm_qcs8550_proxy.zip"
199
- }
200
- },
201
- "qualcomm-qcs9075": {
202
- "precompiled_qnn_onnx": {
203
- "tool_versions": {
204
- "qairt": "2.42.0.251225135753_193295",
205
  "onnx_runtime": "1.25.0"
206
  },
207
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.55.0/gkt-precompiled_qnn_onnx-w8a16_mixed_fp16-qualcomm_qcs9075.zip"
208
  }
209
  }
210
  }
 
1
  {
2
+ "version": "0.56.0",
3
  "precisions": {
4
  "float": {
5
  "chipset_assets": {
6
+ "qualcomm-snapdragon-8gen1": {
7
+ "qnn_context_binary": {
8
+ "tool_versions": {
9
+ "qairt": "2.45.0.260326154327"
10
+ },
11
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-qnn_context_binary-float-qualcomm_snapdragon_8gen1.zip"
12
+ },
13
+ "precompiled_qnn_onnx": {
14
+ "tool_versions": {
15
+ "qairt": "2.45.0.260326154327",
16
+ "onnx_runtime": "1.25.0"
17
+ },
18
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-precompiled_qnn_onnx-float-qualcomm_snapdragon_8gen1.zip"
19
+ }
20
+ },
21
  "qualcomm-snapdragon-8gen3": {
22
  "qnn_context_binary": {
23
  "tool_versions": {
24
  "qairt": "2.45.0.260326154327"
25
  },
26
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-qnn_context_binary-float-qualcomm_snapdragon_8gen3.zip"
27
  },
28
  "precompiled_qnn_onnx": {
29
  "tool_versions": {
30
+ "qairt": "2.45.0.260326154327",
31
  "onnx_runtime": "1.25.0"
32
  },
33
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-precompiled_qnn_onnx-float-qualcomm_snapdragon_8gen3.zip"
34
  }
35
  },
36
  "qualcomm-snapdragon-8-elite-for-galaxy": {
 
38
  "tool_versions": {
39
  "qairt": "2.45.0.260326154327"
40
  },
41
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-qnn_context_binary-float-qualcomm_snapdragon_8_elite_for_galaxy.zip"
42
  },
43
  "precompiled_qnn_onnx": {
44
  "tool_versions": {
45
+ "qairt": "2.45.0.260326154327",
46
  "onnx_runtime": "1.25.0"
47
  },
48
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-precompiled_qnn_onnx-float-qualcomm_snapdragon_8_elite_for_galaxy.zip"
49
  }
50
  },
51
+ "qualcomm-snapdragon-8-elite-gen5-for-galaxy": {
52
  "qnn_context_binary": {
53
  "tool_versions": {
54
  "qairt": "2.45.0.260326154327"
55
  },
56
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-qnn_context_binary-float-qualcomm_snapdragon_8_elite_gen5_for_galaxy.zip"
57
  },
58
  "precompiled_qnn_onnx": {
59
  "tool_versions": {
60
+ "qairt": "2.45.0.260326154327",
61
  "onnx_runtime": "1.25.0"
62
  },
63
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-precompiled_qnn_onnx-float-qualcomm_snapdragon_8_elite_gen5_for_galaxy.zip"
64
  }
65
  },
66
  "qualcomm-snapdragon-x-elite": {
 
68
  "tool_versions": {
69
  "qairt": "2.45.0.260326154327"
70
  },
71
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-qnn_context_binary-float-qualcomm_snapdragon_x_elite.zip"
72
  },
73
  "precompiled_qnn_onnx": {
74
  "tool_versions": {
75
+ "qairt": "2.45.0.260326154327",
76
  "onnx_runtime": "1.25.0"
77
  },
78
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-precompiled_qnn_onnx-float-qualcomm_snapdragon_x_elite.zip"
79
  }
80
  },
81
  "qualcomm-snapdragon-x2-elite": {
 
83
  "tool_versions": {
84
  "qairt": "2.45.0.260326154327"
85
  },
86
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-qnn_context_binary-float-qualcomm_snapdragon_x2_elite.zip"
87
  },
88
  "precompiled_qnn_onnx": {
89
  "tool_versions": {
90
+ "qairt": "2.45.0.260326154327",
91
  "onnx_runtime": "1.25.0"
92
  },
93
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-precompiled_qnn_onnx-float-qualcomm_snapdragon_x2_elite.zip"
94
  }
95
  },
96
  "qualcomm-sa7255p": {
 
98
  "tool_versions": {
99
  "qairt": "2.45.0.260326154327"
100
  },
101
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-qnn_context_binary-float-qualcomm_sa7255p.zip"
102
  }
103
  },
104
  "qualcomm-sa8295p": {
 
106
  "tool_versions": {
107
  "qairt": "2.45.0.260326154327"
108
  },
109
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-qnn_context_binary-float-qualcomm_sa8295p.zip"
110
  }
111
  },
112
  "qualcomm-sa8775p": {
 
114
  "tool_versions": {
115
  "qairt": "2.45.0.260326154327"
116
  },
117
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-qnn_context_binary-float-qualcomm_sa8775p.zip"
118
  }
119
  },
120
  "qualcomm-qcs8550-proxy": {
 
122
  "tool_versions": {
123
  "qairt": "2.45.0.260326154327"
124
  },
125
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-qnn_context_binary-float-qualcomm_qcs8550_proxy.zip"
126
  },
127
  "precompiled_qnn_onnx": {
128
  "tool_versions": {
129
+ "qairt": "2.45.0.260326154327",
130
  "onnx_runtime": "1.25.0"
131
  },
132
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-precompiled_qnn_onnx-float-qualcomm_qcs8550_proxy.zip"
133
  }
134
  },
135
  "qualcomm-qcs9075": {
 
137
  "tool_versions": {
138
  "qairt": "2.45.0.260326154327"
139
  },
140
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-qnn_context_binary-float-qualcomm_qcs9075.zip"
141
  },
142
  "precompiled_qnn_onnx": {
143
  "tool_versions": {
144
+ "qairt": "2.45.0.260326154327",
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
145
  "onnx_runtime": "1.25.0"
146
  },
147
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/gkt/releases/v0.56.0/gkt-precompiled_qnn_onnx-float-qualcomm_qcs9075.zip"
148
  }
149
  }
150
  }