v0.65.0
Browse filesSee https://github.com/qualcomm/ai-hub-models/releases/v0.65.0 for changelog.
- README.md +86 -100
- release_assets.json +8 -13
README.md
CHANGED
|
@@ -16,7 +16,7 @@ pipeline_tag: text-generation
|
|
| 16 |
All-MiniLM-L6-v2 maps sentences to a 384-dimensional dense vector space. Trained on 1B+ sentence pairs, it excels at semantic search, clustering, and sentence similarity tasks while being small enough to run on mobile devices.
|
| 17 |
|
| 18 |
This is based on the implementation of MiniLM-v2 found [here](https://github.com/UKPLab/sentence-transformers).
|
| 19 |
-
This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.
|
| 20 |
|
| 21 |
Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) to run these models on a hosted Qualcomm® device.
|
| 22 |
|
|
@@ -29,26 +29,25 @@ Below are pre-exported model assets ready for deployment.
|
|
| 29 |
|
| 30 |
| Runtime | Precision | Chipset | SDK Versions | Download |
|
| 31 |
|---|---|---|---|---|
|
| 32 |
-
| ONNX | float | Universal | QAIRT 2.50, ONNX Runtime 1.30.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/minilm_v2/releases/v0.
|
| 33 |
-
| ONNX | w8a8 | Universal | QAIRT 2.50, ONNX Runtime 1.30.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/minilm_v2/releases/v0.
|
| 34 |
-
| QNN_DLC | float | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/minilm_v2/releases/v0.
|
| 35 |
-
| QNN_DLC | w8a8 | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/minilm_v2/releases/v0.
|
| 36 |
-
| TFLITE | float | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/minilm_v2/releases/v0.
|
| 37 |
-
| TFLITE | w8a8 | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/minilm_v2/releases/v0.64.0/minilm_v2-tflite-w8a8.zip)
|
| 38 |
|
| 39 |
For more device-specific assets and performance metrics, visit **[MiniLM-v2 on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/minilm_v2)**.
|
| 40 |
|
| 41 |
|
| 42 |
### Option 2: Export with Custom Configurations
|
| 43 |
|
| 44 |
-
Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.
|
| 45 |
- Custom weights (e.g., fine-tuned checkpoints)
|
| 46 |
- Custom input shapes
|
| 47 |
- Target device and runtime configurations
|
| 48 |
|
| 49 |
This option is ideal if you need to customize the model beyond the default configuration provided here.
|
| 50 |
|
| 51 |
-
See our repository for [MiniLM-v2 on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.
|
| 52 |
|
| 53 |
## Model Details
|
| 54 |
|
|
@@ -64,100 +63,87 @@ See our repository for [MiniLM-v2 on GitHub](https://github.com/qualcomm/ai-hub-
|
|
| 64 |
## Performance Summary
|
| 65 |
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|
| 66 |
|---|---|---|---|---|---|---
|
| 67 |
-
| MiniLM-v2 | ONNX | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 0.
|
| 68 |
-
| MiniLM-v2 | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.
|
| 69 |
-
| MiniLM-v2 | ONNX | float | Snapdragon® X2 Elite | 0.
|
| 70 |
-
| MiniLM-v2 | ONNX | float | Snapdragon® X Elite | 1.
|
| 71 |
| MiniLM-v2 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 0.743 ms | 0 - 104 MB | NPU
|
| 72 |
-
| MiniLM-v2 | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 1.812 ms | 0 -
|
| 73 |
-
| MiniLM-v2 | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 1.
|
| 74 |
-
| MiniLM-v2 | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.
|
| 75 |
-
| MiniLM-v2 | ONNX | float | Qualcomm® QCS8450 | 1.812 ms | 0 -
|
| 76 |
-
| MiniLM-v2 | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 1.
|
| 77 |
-
| MiniLM-v2 | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 1.
|
| 78 |
-
| MiniLM-v2 | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 0.
|
| 79 |
-
| MiniLM-v2 | ONNX |
|
| 80 |
-
| MiniLM-v2 | ONNX |
|
| 81 |
-
| MiniLM-v2 | ONNX | w8a8 | Snapdragon®
|
| 82 |
-
| MiniLM-v2 | ONNX | w8a8 | Snapdragon®
|
| 83 |
-
| MiniLM-v2 | ONNX | w8a8 | Snapdragon®
|
| 84 |
-
| MiniLM-v2 | ONNX | w8a8 | Snapdragon®
|
| 85 |
-
| MiniLM-v2 | ONNX | w8a8 |
|
| 86 |
-
| MiniLM-v2 | ONNX | w8a8 | Qualcomm® Dragonwing™
|
| 87 |
-
| MiniLM-v2 | ONNX | w8a8 | Qualcomm® Dragonwing™
|
| 88 |
-
| MiniLM-v2 | ONNX | w8a8 | Qualcomm®
|
| 89 |
-
| MiniLM-v2 | ONNX | w8a8 | Qualcomm®
|
| 90 |
-
| MiniLM-v2 | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-
|
| 91 |
-
| MiniLM-v2 | ONNX | w8a8 | Qualcomm® Dragonwing™
|
| 92 |
-
| MiniLM-v2 | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-
|
| 93 |
-
| MiniLM-v2 | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-
|
| 94 |
-
| MiniLM-v2 | ONNX | w8a8 |
|
| 95 |
-
| MiniLM-v2 |
|
| 96 |
-
| MiniLM-v2 |
|
| 97 |
-
| MiniLM-v2 |
|
| 98 |
-
| MiniLM-v2 | QNN_DLC | float | Snapdragon®
|
|
|
|
|
|
|
|
|
|
| 99 |
| MiniLM-v2 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 0.733 ms | 0 - 104 MB | NPU
|
| 100 |
-
| MiniLM-v2 | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 1.
|
| 101 |
-
| MiniLM-v2 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 1.
|
| 102 |
-
| MiniLM-v2 | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.
|
| 103 |
-
| MiniLM-v2 | QNN_DLC | float | Qualcomm® SA8775P | 1.
|
| 104 |
-
| MiniLM-v2 | QNN_DLC | float | Qualcomm® SA8650P | 1.
|
| 105 |
-
| MiniLM-v2 | QNN_DLC | float | Qualcomm® SA8255P | 1.
|
| 106 |
-
| MiniLM-v2 | QNN_DLC | float | Qualcomm® QCS8450 | 1.
|
| 107 |
-
| MiniLM-v2 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 1.
|
| 108 |
-
| MiniLM-v2 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 1.
|
| 109 |
-
| MiniLM-v2 | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 0.
|
| 110 |
-
| MiniLM-v2 | QNN_DLC | float | Qualcomm® SA8295P | 2.
|
| 111 |
-
| MiniLM-v2 | QNN_DLC |
|
| 112 |
-
| MiniLM-v2 | QNN_DLC |
|
| 113 |
-
| MiniLM-v2 | QNN_DLC | w8a8 | Snapdragon®
|
| 114 |
-
| MiniLM-v2 | QNN_DLC | w8a8 | Snapdragon®
|
| 115 |
-
| MiniLM-v2 | QNN_DLC | w8a8 | Snapdragon®
|
| 116 |
-
| MiniLM-v2 | QNN_DLC | w8a8 | Snapdragon®
|
| 117 |
-
| MiniLM-v2 | QNN_DLC | w8a8 |
|
| 118 |
-
| MiniLM-v2 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™
|
| 119 |
-
| MiniLM-v2 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™
|
| 120 |
-
| MiniLM-v2 | QNN_DLC | w8a8 | Qualcomm®
|
| 121 |
-
| MiniLM-v2 | QNN_DLC | w8a8 | Qualcomm®
|
| 122 |
-
| MiniLM-v2 | QNN_DLC | w8a8 | Qualcomm®
|
| 123 |
-
| MiniLM-v2 | QNN_DLC | w8a8 | Qualcomm®
|
| 124 |
-
| MiniLM-v2 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™
|
| 125 |
-
| MiniLM-v2 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™
|
| 126 |
-
| MiniLM-v2 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-
|
| 127 |
-
| MiniLM-v2 | QNN_DLC | w8a8 | Qualcomm®
|
| 128 |
-
| MiniLM-v2 | QNN_DLC | w8a8 | Qualcomm®
|
| 129 |
-
| MiniLM-v2 | QNN_DLC | w8a8 |
|
| 130 |
-
| MiniLM-v2 | QNN_DLC | w8a8 | Snapdragon®
|
| 131 |
-
| MiniLM-v2 |
|
| 132 |
-
| MiniLM-v2 | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.
|
|
|
|
| 133 |
| MiniLM-v2 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 0.772 ms | 0 - 107 MB | NPU
|
| 134 |
-
| MiniLM-v2 | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 1.
|
| 135 |
-
| MiniLM-v2 | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 1.
|
| 136 |
-
| MiniLM-v2 | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.
|
| 137 |
-
| MiniLM-v2 | TFLITE | float | Qualcomm®
|
| 138 |
-
| MiniLM-v2 | TFLITE | float | Qualcomm®
|
| 139 |
-
| MiniLM-v2 | TFLITE | float | Qualcomm®
|
| 140 |
-
| MiniLM-v2 | TFLITE | float | Qualcomm®
|
| 141 |
-
| MiniLM-v2 | TFLITE | float | Qualcomm®
|
| 142 |
-
| MiniLM-v2 | TFLITE | float |
|
| 143 |
-
| MiniLM-v2 | TFLITE | float |
|
| 144 |
-
| MiniLM-v2 | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 0.849 ms | 0 - 56 MB | NPU
|
| 145 |
-
| MiniLM-v2 | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 1.074 ms | 0 - 57 MB | NPU
|
| 146 |
-
| MiniLM-v2 | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 1.524 ms | 0 - 77 MB | NPU
|
| 147 |
-
| MiniLM-v2 | TFLITE | w8a8 | Snapdragon® 8 Gen 1 Mobile | 2.352 ms | 0 - 84 MB | NPU
|
| 148 |
-
| MiniLM-v2 | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 16.469 ms | 0 - 32 MB | NPU
|
| 149 |
-
| MiniLM-v2 | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 2.065 ms | 0 - 25 MB | NPU
|
| 150 |
-
| MiniLM-v2 | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 2.151 ms | 0 - 2 MB | NPU
|
| 151 |
-
| MiniLM-v2 | TFLITE | w8a8 | Qualcomm® SA8775P | 2.415 ms | 0 - 61 MB | NPU
|
| 152 |
-
| MiniLM-v2 | TFLITE | w8a8 | Qualcomm® SA8650P | 2.415 ms | 0 - 61 MB | NPU
|
| 153 |
-
| MiniLM-v2 | TFLITE | w8a8 | Qualcomm® SA8255P | 2.415 ms | 0 - 61 MB | NPU
|
| 154 |
-
| MiniLM-v2 | TFLITE | w8a8 | Qualcomm® QCS8450 | 2.352 ms | 0 - 84 MB | NPU
|
| 155 |
-
| MiniLM-v2 | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 2.28 ms | 0 - 24 MB | NPU
|
| 156 |
-
| MiniLM-v2 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 12.133 ms | 0 - 50 MB | NPU
|
| 157 |
-
| MiniLM-v2 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 5.375 ms | 0 - 38 MB | NPU
|
| 158 |
-
| MiniLM-v2 | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 1.074 ms | 0 - 57 MB | NPU
|
| 159 |
-
| MiniLM-v2 | TFLITE | w8a8 | Qualcomm® SA8295P | 2.848 ms | 0 - 52 MB | NPU
|
| 160 |
-
| MiniLM-v2 | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 5.375 ms | 0 - 38 MB | NPU
|
| 161 |
|
| 162 |
## License
|
| 163 |
* The license for the original implementation of MiniLM-v2 can be found
|
|
|
|
| 16 |
All-MiniLM-L6-v2 maps sentences to a 384-dimensional dense vector space. Trained on 1B+ sentence pairs, it excels at semantic search, clustering, and sentence similarity tasks while being small enough to run on mobile devices.
|
| 17 |
|
| 18 |
This is based on the implementation of MiniLM-v2 found [here](https://github.com/UKPLab/sentence-transformers).
|
| 19 |
+
This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.65.0/src/qai_hub_models/models/minilm_v2) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
|
| 20 |
|
| 21 |
Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) to run these models on a hosted Qualcomm® device.
|
| 22 |
|
|
|
|
| 29 |
|
| 30 |
| Runtime | Precision | Chipset | SDK Versions | Download |
|
| 31 |
|---|---|---|---|---|
|
| 32 |
+
| ONNX | float | Universal | QAIRT 2.50, ONNX Runtime 1.30.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/minilm_v2/releases/v0.65.0/minilm_v2-onnx-float.zip)
|
| 33 |
+
| ONNX | w8a8 | Universal | QAIRT 2.50, ONNX Runtime 1.30.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/minilm_v2/releases/v0.65.0/minilm_v2-onnx-w8a8.zip)
|
| 34 |
+
| QNN_DLC | float | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/minilm_v2/releases/v0.65.0/minilm_v2-qnn_dlc-float.zip)
|
| 35 |
+
| QNN_DLC | w8a8 | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/minilm_v2/releases/v0.65.0/minilm_v2-qnn_dlc-w8a8.zip)
|
| 36 |
+
| TFLITE | float | Universal | QAIRT 2.50 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/minilm_v2/releases/v0.65.0/minilm_v2-tflite-float.zip)
|
|
|
|
| 37 |
|
| 38 |
For more device-specific assets and performance metrics, visit **[MiniLM-v2 on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/minilm_v2)**.
|
| 39 |
|
| 40 |
|
| 41 |
### Option 2: Export with Custom Configurations
|
| 42 |
|
| 43 |
+
Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.65.0/src/qai_hub_models/models/minilm_v2) Python library to compile and export the model with your own:
|
| 44 |
- Custom weights (e.g., fine-tuned checkpoints)
|
| 45 |
- Custom input shapes
|
| 46 |
- Target device and runtime configurations
|
| 47 |
|
| 48 |
This option is ideal if you need to customize the model beyond the default configuration provided here.
|
| 49 |
|
| 50 |
+
See our repository for [MiniLM-v2 on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.65.0/src/qai_hub_models/models/minilm_v2) for usage instructions.
|
| 51 |
|
| 52 |
## Model Details
|
| 53 |
|
|
|
|
| 63 |
## Performance Summary
|
| 64 |
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|
| 65 |
|---|---|---|---|---|---|---
|
| 66 |
+
| MiniLM-v2 | ONNX | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 0.401 ms | 0 - 65 MB | NPU
|
| 67 |
+
| MiniLM-v2 | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.504 ms | 0 - 65 MB | NPU
|
| 68 |
+
| MiniLM-v2 | ONNX | float | Snapdragon® X2 Elite | 0.421 ms | 1 - 1 MB | NPU
|
| 69 |
+
| MiniLM-v2 | ONNX | float | Snapdragon® X Elite | 1.066 ms | 45 - 45 MB | NPU
|
| 70 |
| MiniLM-v2 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 0.743 ms | 0 - 104 MB | NPU
|
| 71 |
+
| MiniLM-v2 | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 1.812 ms | 0 - 106 MB | NPU
|
| 72 |
+
| MiniLM-v2 | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 1.293 ms | 0 - 4 MB | NPU
|
| 73 |
+
| MiniLM-v2 | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.049 ms | 0 - 12 MB | NPU
|
| 74 |
+
| MiniLM-v2 | ONNX | float | Qualcomm® QCS8450 | 1.812 ms | 0 - 106 MB | NPU
|
| 75 |
+
| MiniLM-v2 | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 1.303 ms | 0 - 3 MB | NPU
|
| 76 |
+
| MiniLM-v2 | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 1.066 ms | 45 - 45 MB | NPU
|
| 77 |
+
| MiniLM-v2 | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 0.504 ms | 0 - 65 MB | NPU
|
| 78 |
+
| MiniLM-v2 | ONNX | float | Snapdragon® 8 Elite Extreme Gen 6 Mobile | 0.336 ms | 0 - 64 MB | NPU
|
| 79 |
+
| MiniLM-v2 | ONNX | float | Snapdragon® 8 Elite Gen 6 Mobile | 0.443 ms | 0 - 65 MB | NPU
|
| 80 |
+
| MiniLM-v2 | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 0.667 ms | 0 - 63 MB | NPU
|
| 81 |
+
| MiniLM-v2 | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.845 ms | 0 - 61 MB | NPU
|
| 82 |
+
| MiniLM-v2 | ONNX | w8a8 | Snapdragon® X2 Elite | 0.668 ms | 1 - 1 MB | NPU
|
| 83 |
+
| MiniLM-v2 | ONNX | w8a8 | Snapdragon® X Elite | 1.797 ms | 22 - 22 MB | NPU
|
| 84 |
+
| MiniLM-v2 | ONNX | w8a8 | Snapdragon® 8 Gen 1 Mobile | 1.94 ms | 0 - 89 MB | NPU
|
| 85 |
+
| MiniLM-v2 | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 4.455 ms | 0 - 3 MB | NPU
|
| 86 |
+
| MiniLM-v2 | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 1.691 ms | 0 - 4 MB | NPU
|
| 87 |
+
| MiniLM-v2 | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.719 ms | 0 - 2 MB | NPU
|
| 88 |
+
| MiniLM-v2 | ONNX | w8a8 | Qualcomm® QCS8450 | 1.94 ms | 0 - 89 MB | NPU
|
| 89 |
+
| MiniLM-v2 | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 1.834 ms | 0 - 3 MB | NPU
|
| 90 |
+
| MiniLM-v2 | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 1.797 ms | 22 - 22 MB | NPU
|
| 91 |
+
| MiniLM-v2 | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 6.44 ms | 0 - 64 MB | NPU
|
| 92 |
+
| MiniLM-v2 | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 1.784 ms | 0 - 56 MB | NPU
|
| 93 |
+
| MiniLM-v2 | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.845 ms | 0 - 61 MB | NPU
|
| 94 |
+
| MiniLM-v2 | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.784 ms | 0 - 56 MB | NPU
|
| 95 |
+
| MiniLM-v2 | ONNX | w8a8 | Snapdragon® 8 Elite Extreme Gen 6 Mobile | 0.436 ms | 0 - 63 MB | NPU
|
| 96 |
+
| MiniLM-v2 | ONNX | w8a8 | Snapdragon® 8 Elite Gen 6 Mobile | 0.686 ms | 0 - 64 MB | NPU
|
| 97 |
+
| MiniLM-v2 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 0.407 ms | 0 - 62 MB | NPU
|
| 98 |
+
| MiniLM-v2 | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.506 ms | 0 - 62 MB | NPU
|
| 99 |
+
| MiniLM-v2 | QNN_DLC | float | Snapdragon® X2 Elite | 0.604 ms | 1 - 1 MB | NPU
|
| 100 |
+
| MiniLM-v2 | QNN_DLC | float | Snapdragon® X Elite | 1.218 ms | 1 - 1 MB | NPU
|
| 101 |
| MiniLM-v2 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 0.733 ms | 0 - 104 MB | NPU
|
| 102 |
+
| MiniLM-v2 | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 1.806 ms | 0 - 104 MB | NPU
|
| 103 |
+
| MiniLM-v2 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 1.259 ms | 0 - 3 MB | NPU
|
| 104 |
+
| MiniLM-v2 | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.041 ms | 0 - 2 MB | NPU
|
| 105 |
+
| MiniLM-v2 | QNN_DLC | float | Qualcomm® SA8775P | 1.414 ms | 0 - 63 MB | NPU
|
| 106 |
+
| MiniLM-v2 | QNN_DLC | float | Qualcomm® SA8650P | 1.414 ms | 0 - 63 MB | NPU
|
| 107 |
+
| MiniLM-v2 | QNN_DLC | float | Qualcomm® SA8255P | 1.414 ms | 0 - 63 MB | NPU
|
| 108 |
+
| MiniLM-v2 | QNN_DLC | float | Qualcomm® QCS8450 | 1.806 ms | 0 - 104 MB | NPU
|
| 109 |
+
| MiniLM-v2 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 1.3 ms | 2 - 4 MB | NPU
|
| 110 |
+
| MiniLM-v2 | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 1.218 ms | 1 - 1 MB | NPU
|
| 111 |
+
| MiniLM-v2 | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 0.506 ms | 0 - 62 MB | NPU
|
| 112 |
+
| MiniLM-v2 | QNN_DLC | float | Qualcomm® SA8295P | 2.077 ms | 0 - 60 MB | NPU
|
| 113 |
+
| MiniLM-v2 | QNN_DLC | float | Snapdragon® 8 Elite Extreme Gen 6 Mobile | 0.323 ms | 0 - 64 MB | NPU
|
| 114 |
+
| MiniLM-v2 | QNN_DLC | float | Snapdragon® 8 Elite Gen 6 Mobile | 0.44 ms | 0 - 65 MB | NPU
|
| 115 |
+
| MiniLM-v2 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 0.688 ms | 0 - 56 MB | NPU
|
| 116 |
+
| MiniLM-v2 | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.854 ms | 0 - 58 MB | NPU
|
| 117 |
+
| MiniLM-v2 | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.813 ms | 0 - 0 MB | NPU
|
| 118 |
+
| MiniLM-v2 | QNN_DLC | w8a8 | Snapdragon® X Elite | 1.963 ms | 1 - 1 MB | NPU
|
| 119 |
+
| MiniLM-v2 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 1 Mobile | 1.963 ms | 0 - 79 MB | NPU
|
| 120 |
+
| MiniLM-v2 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 3.818 ms | 2 - 4 MB | NPU
|
| 121 |
+
| MiniLM-v2 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-8275 | 1.68 ms | 0 - 3 MB | NPU
|
| 122 |
+
| MiniLM-v2 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.747 ms | 0 - 2 MB | NPU
|
| 123 |
+
| MiniLM-v2 | QNN_DLC | w8a8 | Qualcomm® QCS8450 | 1.963 ms | 0 - 79 MB | NPU
|
| 124 |
+
| MiniLM-v2 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 1.851 ms | 0 - 2 MB | NPU
|
| 125 |
+
| MiniLM-v2 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 1.963 ms | 1 - 1 MB | NPU
|
| 126 |
+
| MiniLM-v2 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 6.238 ms | 0 - 55 MB | NPU
|
| 127 |
+
| MiniLM-v2 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 1.733 ms | 0 - 52 MB | NPU
|
| 128 |
+
| MiniLM-v2 | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.854 ms | 0 - 58 MB | NPU
|
| 129 |
+
| MiniLM-v2 | QNN_DLC | w8a8 | Qualcomm® SA7255P | 3.419 ms | 0 - 54 MB | NPU
|
| 130 |
+
| MiniLM-v2 | QNN_DLC | w8a8 | Qualcomm® SA8295P | 2.312 ms | 0 - 52 MB | NPU
|
| 131 |
+
| MiniLM-v2 | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.733 ms | 0 - 52 MB | NPU
|
| 132 |
+
| MiniLM-v2 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Extreme Gen 6 Mobile | 0.445 ms | 0 - 60 MB | NPU
|
| 133 |
+
| MiniLM-v2 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 6 Mobile | 0.692 ms | 0 - 59 MB | NPU
|
| 134 |
+
| MiniLM-v2 | TFLITE | float | Snapdragon® 8 Elite Gen 5 For Galaxy Mobile | 0.436 ms | 0 - 59 MB | NPU
|
| 135 |
+
| MiniLM-v2 | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.537 ms | 0 - 66 MB | NPU
|
| 136 |
| MiniLM-v2 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 0.772 ms | 0 - 107 MB | NPU
|
| 137 |
+
| MiniLM-v2 | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 1.873 ms | 0 - 106 MB | NPU
|
| 138 |
+
| MiniLM-v2 | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 1.313 ms | 0 - 47 MB | NPU
|
| 139 |
+
| MiniLM-v2 | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.054 ms | 0 - 2 MB | NPU
|
| 140 |
+
| MiniLM-v2 | TFLITE | float | Qualcomm® QCS8450 | 1.873 ms | 0 - 106 MB | NPU
|
| 141 |
+
| MiniLM-v2 | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 1.329 ms | 0 - 46 MB | NPU
|
| 142 |
+
| MiniLM-v2 | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 0.537 ms | 0 - 66 MB | NPU
|
| 143 |
+
| MiniLM-v2 | TFLITE | float | Qualcomm® SA7255P | 3.581 ms | 0 - 65 MB | NPU
|
| 144 |
+
| MiniLM-v2 | TFLITE | float | Qualcomm® SA8295P | 2.159 ms | 0 - 59 MB | NPU
|
| 145 |
+
| MiniLM-v2 | TFLITE | float | Snapdragon® 8 Elite Extreme Gen 6 Mobile | 0.348 ms | 0 - 69 MB | NPU
|
| 146 |
+
| MiniLM-v2 | TFLITE | float | Snapdragon® 8 Elite Gen 6 Mobile | 0.467 ms | 0 - 70 MB | NPU
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 147 |
|
| 148 |
## License
|
| 149 |
* The license for the original implementation of MiniLM-v2 can be found
|
release_assets.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"version": "0.
|
| 3 |
"precisions": {
|
| 4 |
"float": {
|
| 5 |
"universal_assets": {
|
|
@@ -8,19 +8,20 @@
|
|
| 8 |
"qairt": "2.50.0.260828221209",
|
| 9 |
"onnx_runtime": "1.30.0"
|
| 10 |
},
|
| 11 |
-
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/minilm_v2/releases/v0.
|
| 12 |
},
|
| 13 |
"qnn_dlc": {
|
| 14 |
"tool_versions": {
|
| 15 |
"qairt": "2.50.0.260828221209"
|
| 16 |
},
|
| 17 |
-
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/minilm_v2/releases/v0.
|
| 18 |
},
|
| 19 |
"tflite": {
|
| 20 |
"tool_versions": {
|
| 21 |
-
"qairt": "2.50.0.260828221209"
|
|
|
|
| 22 |
},
|
| 23 |
-
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/minilm_v2/releases/v0.
|
| 24 |
}
|
| 25 |
}
|
| 26 |
},
|
|
@@ -31,19 +32,13 @@
|
|
| 31 |
"qairt": "2.50.0.260828221209",
|
| 32 |
"onnx_runtime": "1.30.0"
|
| 33 |
},
|
| 34 |
-
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/minilm_v2/releases/v0.
|
| 35 |
},
|
| 36 |
"qnn_dlc": {
|
| 37 |
"tool_versions": {
|
| 38 |
"qairt": "2.50.0.260828221209"
|
| 39 |
},
|
| 40 |
-
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/minilm_v2/releases/v0.
|
| 41 |
-
},
|
| 42 |
-
"tflite": {
|
| 43 |
-
"tool_versions": {
|
| 44 |
-
"qairt": "2.50.0.260828221209"
|
| 45 |
-
},
|
| 46 |
-
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/minilm_v2/releases/v0.64.0/minilm_v2-tflite-w8a8.zip"
|
| 47 |
}
|
| 48 |
}
|
| 49 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"version": "0.65.0",
|
| 3 |
"precisions": {
|
| 4 |
"float": {
|
| 5 |
"universal_assets": {
|
|
|
|
| 8 |
"qairt": "2.50.0.260828221209",
|
| 9 |
"onnx_runtime": "1.30.0"
|
| 10 |
},
|
| 11 |
+
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/minilm_v2/releases/v0.65.0/minilm_v2-onnx-float.zip"
|
| 12 |
},
|
| 13 |
"qnn_dlc": {
|
| 14 |
"tool_versions": {
|
| 15 |
"qairt": "2.50.0.260828221209"
|
| 16 |
},
|
| 17 |
+
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/minilm_v2/releases/v0.65.0/minilm_v2-qnn_dlc-float.zip"
|
| 18 |
},
|
| 19 |
"tflite": {
|
| 20 |
"tool_versions": {
|
| 21 |
+
"qairt": "2.50.0.260828221209",
|
| 22 |
+
"litert": "1.4.2"
|
| 23 |
},
|
| 24 |
+
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/minilm_v2/releases/v0.65.0/minilm_v2-tflite-float.zip"
|
| 25 |
}
|
| 26 |
}
|
| 27 |
},
|
|
|
|
| 32 |
"qairt": "2.50.0.260828221209",
|
| 33 |
"onnx_runtime": "1.30.0"
|
| 34 |
},
|
| 35 |
+
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/minilm_v2/releases/v0.65.0/minilm_v2-onnx-w8a8.zip"
|
| 36 |
},
|
| 37 |
"qnn_dlc": {
|
| 38 |
"tool_versions": {
|
| 39 |
"qairt": "2.50.0.260828221209"
|
| 40 |
},
|
| 41 |
+
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/minilm_v2/releases/v0.65.0/minilm_v2-qnn_dlc-w8a8.zip"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 42 |
}
|
| 43 |
}
|
| 44 |
}
|