v0.31.0
Browse filesSee https://github.com/quic/ai-hub-models/releases/v0.31.0 for changelog.
README.md
CHANGED
|
@@ -36,51 +36,51 @@ More details on model performance across various devices, can be found
|
|
| 36 |
|
| 37 |
| Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
|
| 38 |
|---|---|---|---|---|---|---|---|---|
|
| 39 |
-
| text_encoder | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) |
|
| 40 |
-
| text_encoder | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) |
|
| 41 |
-
| text_encoder | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) |
|
| 42 |
-
| text_encoder | w8a16 | SA7255P ADP | Qualcomm® SA7255P |
|
| 43 |
-
| text_encoder | w8a16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) |
|
| 44 |
-
| text_encoder | w8a16 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) |
|
| 45 |
-
| text_encoder | w8a16 | SA8775P ADP | Qualcomm® SA8775P |
|
| 46 |
-
| text_encoder | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile |
|
| 47 |
-
| text_encoder | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile |
|
| 48 |
-
| text_encoder | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile |
|
| 49 |
-
| text_encoder | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile |
|
| 50 |
-
| text_encoder | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile |
|
| 51 |
-
| text_encoder | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile |
|
| 52 |
-
| text_encoder | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite |
|
| 53 |
-
| text_encoder | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite |
|
| 54 |
-
| unet | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) |
|
| 55 |
-
| unet | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) |
|
| 56 |
-
| unet | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) |
|
| 57 |
-
| unet | w8a16 | SA7255P ADP | Qualcomm® SA7255P |
|
| 58 |
-
| unet | w8a16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) |
|
| 59 |
-
| unet | w8a16 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) |
|
| 60 |
-
| unet | w8a16 | SA8775P ADP | Qualcomm® SA8775P |
|
| 61 |
-
| unet | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile |
|
| 62 |
-
| unet | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile |
|
| 63 |
-
| unet | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile |
|
| 64 |
-
| unet | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile |
|
| 65 |
-
| unet | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile |
|
| 66 |
-
| unet | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile |
|
| 67 |
-
| unet | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite |
|
| 68 |
-
| unet | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite |
|
| 69 |
-
| vae | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) |
|
| 70 |
-
| vae | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) |
|
| 71 |
-
| vae | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) |
|
| 72 |
-
| vae | w8a16 | SA7255P ADP | Qualcomm® SA7255P |
|
| 73 |
-
| vae | w8a16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) |
|
| 74 |
-
| vae | w8a16 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) |
|
| 75 |
-
| vae | w8a16 | SA8775P ADP | Qualcomm® SA8775P |
|
| 76 |
-
| vae | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile |
|
| 77 |
-
| vae | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile |
|
| 78 |
-
| vae | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile |
|
| 79 |
-
| vae | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile |
|
| 80 |
-
| vae | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile |
|
| 81 |
-
| vae | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile |
|
| 82 |
-
| vae | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite |
|
| 83 |
-
| vae | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite |
|
| 84 |
|
| 85 |
## Deploy to Snapdragon X Elite NPU
|
| 86 |
Please follow the [Stable Diffusion Windows App](https://github.com/quic/ai-hub-apps/tree/main/apps/windows/python/StableDiffusion) tutorial to quantize model with custom weights.
|
|
@@ -149,8 +149,8 @@ Profiling Results
|
|
| 149 |
------------------------------------------------------------
|
| 150 |
text_encoder
|
| 151 |
Device : cs_8275 (ANDROID 14)
|
| 152 |
-
Runtime :
|
| 153 |
-
Estimated inference time (ms) : 18.
|
| 154 |
Estimated peak memory usage (MB): [0, 9]
|
| 155 |
Total # Ops : 787
|
| 156 |
Compute Unit(s) : npu (787 ops) gpu (0 ops) cpu (0 ops)
|
|
@@ -158,17 +158,17 @@ Compute Unit(s) : npu (787 ops) gpu (0 ops) cpu (0 ops)
|
|
| 158 |
------------------------------------------------------------
|
| 159 |
unet
|
| 160 |
Device : cs_8275 (ANDROID 14)
|
| 161 |
-
Runtime :
|
| 162 |
-
Estimated inference time (ms) : 232.
|
| 163 |
-
Estimated peak memory usage (MB): [
|
| 164 |
Total # Ops : 5783
|
| 165 |
Compute Unit(s) : npu (5783 ops) gpu (0 ops) cpu (0 ops)
|
| 166 |
|
| 167 |
------------------------------------------------------------
|
| 168 |
vae
|
| 169 |
Device : cs_8275 (ANDROID 14)
|
| 170 |
-
Runtime :
|
| 171 |
-
Estimated inference time (ms) :
|
| 172 |
Estimated peak memory usage (MB): [0, 9]
|
| 173 |
Total # Ops : 173
|
| 174 |
Compute Unit(s) : npu (173 ops) gpu (0 ops) cpu (0 ops)
|
|
|
|
| 36 |
|
| 37 |
| Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
|
| 38 |
|---|---|---|---|---|---|---|---|---|
|
| 39 |
+
| text_encoder | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_CONTEXT_BINARY | 18.565 ms | 0 - 9 MB | NPU | Use Export Script |
|
| 40 |
+
| text_encoder | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_CONTEXT_BINARY | 8.2 ms | 0 - 3 MB | NPU | Use Export Script |
|
| 41 |
+
| text_encoder | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_CONTEXT_BINARY | 8.388 ms | 0 - 9 MB | NPU | Use Export Script |
|
| 42 |
+
| text_encoder | w8a16 | SA7255P ADP | Qualcomm® SA7255P | QNN_CONTEXT_BINARY | 18.565 ms | 0 - 9 MB | NPU | Use Export Script |
|
| 43 |
+
| text_encoder | w8a16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_CONTEXT_BINARY | 8.354 ms | 0 - 2 MB | NPU | Use Export Script |
|
| 44 |
+
| text_encoder | w8a16 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_CONTEXT_BINARY | 8.198 ms | 0 - 2 MB | NPU | Use Export Script |
|
| 45 |
+
| text_encoder | w8a16 | SA8775P ADP | Qualcomm® SA8775P | QNN_CONTEXT_BINARY | 8.388 ms | 0 - 9 MB | NPU | Use Export Script |
|
| 46 |
+
| text_encoder | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN_CONTEXT_BINARY | 8.231 ms | 0 - 2 MB | NPU | Use Export Script |
|
| 47 |
+
| text_encoder | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | PRECOMPILED_QNN_ONNX | 8.146 ms | 0 - 386 MB | NPU | Use Export Script |
|
| 48 |
+
| text_encoder | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_CONTEXT_BINARY | 5.736 ms | 0 - 19 MB | NPU | Use Export Script |
|
| 49 |
+
| text_encoder | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | PRECOMPILED_QNN_ONNX | 5.389 ms | 0 - 19 MB | NPU | Use Export Script |
|
| 50 |
+
| text_encoder | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN_CONTEXT_BINARY | 5.126 ms | 0 - 14 MB | NPU | Use Export Script |
|
| 51 |
+
| text_encoder | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | PRECOMPILED_QNN_ONNX | 5.359 ms | 0 - 18 MB | NPU | Use Export Script |
|
| 52 |
+
| text_encoder | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_CONTEXT_BINARY | 8.635 ms | 0 - 0 MB | NPU | Use Export Script |
|
| 53 |
+
| text_encoder | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | PRECOMPILED_QNN_ONNX | 8.267 ms | 378 - 378 MB | NPU | Use Export Script |
|
| 54 |
+
| unet | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_CONTEXT_BINARY | 232.543 ms | 0 - 8 MB | NPU | Use Export Script |
|
| 55 |
+
| unet | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_CONTEXT_BINARY | 93.406 ms | 0 - 3 MB | NPU | Use Export Script |
|
| 56 |
+
| unet | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_CONTEXT_BINARY | 87.879 ms | 0 - 8 MB | NPU | Use Export Script |
|
| 57 |
+
| unet | w8a16 | SA7255P ADP | Qualcomm® SA7255P | QNN_CONTEXT_BINARY | 232.543 ms | 0 - 8 MB | NPU | Use Export Script |
|
| 58 |
+
| unet | w8a16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_CONTEXT_BINARY | 93.629 ms | 0 - 2 MB | NPU | Use Export Script |
|
| 59 |
+
| unet | w8a16 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_CONTEXT_BINARY | 94.226 ms | 0 - 3 MB | NPU | Use Export Script |
|
| 60 |
+
| unet | w8a16 | SA8775P ADP | Qualcomm® SA8775P | QNN_CONTEXT_BINARY | 87.879 ms | 0 - 8 MB | NPU | Use Export Script |
|
| 61 |
+
| unet | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN_CONTEXT_BINARY | 93.9 ms | 0 - 3 MB | NPU | Use Export Script |
|
| 62 |
+
| unet | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | PRECOMPILED_QNN_ONNX | 94.942 ms | 0 - 898 MB | NPU | Use Export Script |
|
| 63 |
+
| unet | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_CONTEXT_BINARY | 65.39 ms | 0 - 17 MB | NPU | Use Export Script |
|
| 64 |
+
| unet | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | PRECOMPILED_QNN_ONNX | 66.62 ms | 0 - 15 MB | NPU | Use Export Script |
|
| 65 |
+
| unet | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN_CONTEXT_BINARY | 58.603 ms | 0 - 15 MB | NPU | Use Export Script |
|
| 66 |
+
| unet | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | PRECOMPILED_QNN_ONNX | 58.424 ms | 0 - 19 MB | NPU | Use Export Script |
|
| 67 |
+
| unet | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_CONTEXT_BINARY | 95.248 ms | 0 - 0 MB | NPU | Use Export Script |
|
| 68 |
+
| unet | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | PRECOMPILED_QNN_ONNX | 96.103 ms | 842 - 842 MB | NPU | Use Export Script |
|
| 69 |
+
| vae | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_CONTEXT_BINARY | 718.787 ms | 0 - 9 MB | NPU | Use Export Script |
|
| 70 |
+
| vae | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_CONTEXT_BINARY | 266.604 ms | 0 - 3 MB | NPU | Use Export Script |
|
| 71 |
+
| vae | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_CONTEXT_BINARY | 249.03 ms | 0 - 10 MB | NPU | Use Export Script |
|
| 72 |
+
| vae | w8a16 | SA7255P ADP | Qualcomm® SA7255P | QNN_CONTEXT_BINARY | 718.787 ms | 0 - 9 MB | NPU | Use Export Script |
|
| 73 |
+
| vae | w8a16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_CONTEXT_BINARY | 268.853 ms | 0 - 2 MB | NPU | Use Export Script |
|
| 74 |
+
| vae | w8a16 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_CONTEXT_BINARY | 267.681 ms | 0 - 2 MB | NPU | Use Export Script |
|
| 75 |
+
| vae | w8a16 | SA8775P ADP | Qualcomm® SA8775P | QNN_CONTEXT_BINARY | 249.03 ms | 0 - 10 MB | NPU | Use Export Script |
|
| 76 |
+
| vae | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN_CONTEXT_BINARY | 271.2 ms | 0 - 3 MB | NPU | Use Export Script |
|
| 77 |
+
| vae | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | PRECOMPILED_QNN_ONNX | 285.73 ms | 0 - 66 MB | NPU | Use Export Script |
|
| 78 |
+
| vae | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_CONTEXT_BINARY | 205.864 ms | 0 - 18 MB | NPU | Use Export Script |
|
| 79 |
+
| vae | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | PRECOMPILED_QNN_ONNX | 205.322 ms | 3 - 21 MB | NPU | Use Export Script |
|
| 80 |
+
| vae | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN_CONTEXT_BINARY | 175.321 ms | 0 - 14 MB | NPU | Use Export Script |
|
| 81 |
+
| vae | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | PRECOMPILED_QNN_ONNX | 194.44 ms | 3 - 17 MB | NPU | Use Export Script |
|
| 82 |
+
| vae | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_CONTEXT_BINARY | 265.17 ms | 0 - 0 MB | NPU | Use Export Script |
|
| 83 |
+
| vae | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | PRECOMPILED_QNN_ONNX | 264.88 ms | 62 - 62 MB | NPU | Use Export Script |
|
| 84 |
|
| 85 |
## Deploy to Snapdragon X Elite NPU
|
| 86 |
Please follow the [Stable Diffusion Windows App](https://github.com/quic/ai-hub-apps/tree/main/apps/windows/python/StableDiffusion) tutorial to quantize model with custom weights.
|
|
|
|
| 149 |
------------------------------------------------------------
|
| 150 |
text_encoder
|
| 151 |
Device : cs_8275 (ANDROID 14)
|
| 152 |
+
Runtime : QNN_CONTEXT_BINARY
|
| 153 |
+
Estimated inference time (ms) : 18.6
|
| 154 |
Estimated peak memory usage (MB): [0, 9]
|
| 155 |
Total # Ops : 787
|
| 156 |
Compute Unit(s) : npu (787 ops) gpu (0 ops) cpu (0 ops)
|
|
|
|
| 158 |
------------------------------------------------------------
|
| 159 |
unet
|
| 160 |
Device : cs_8275 (ANDROID 14)
|
| 161 |
+
Runtime : QNN_CONTEXT_BINARY
|
| 162 |
+
Estimated inference time (ms) : 232.5
|
| 163 |
+
Estimated peak memory usage (MB): [0, 8]
|
| 164 |
Total # Ops : 5783
|
| 165 |
Compute Unit(s) : npu (5783 ops) gpu (0 ops) cpu (0 ops)
|
| 166 |
|
| 167 |
------------------------------------------------------------
|
| 168 |
vae
|
| 169 |
Device : cs_8275 (ANDROID 14)
|
| 170 |
+
Runtime : QNN_CONTEXT_BINARY
|
| 171 |
+
Estimated inference time (ms) : 718.8
|
| 172 |
Estimated peak memory usage (MB): [0, 9]
|
| 173 |
Total # Ops : 173
|
| 174 |
Compute Unit(s) : npu (173 ops) gpu (0 ops) cpu (0 ops)
|