v0.42.0
Browse filesSee https://github.com/quic/ai-hub-models/releases/v0.42.0 for changelog.
- QuickSRNetMedium_float.dlc +2 -2
- QuickSRNetMedium_float.onnx.zip +1 -1
- QuickSRNetMedium_w8a8.dlc +2 -2
- README.md +63 -60
- tool-versions.yaml +1 -1
QuickSRNetMedium_float.dlc
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5b187262ff98d616799c8191c6fa2dcab809ab79aa762644141375137c1fa4d2
|
| 3 |
+
size 274388
|
QuickSRNetMedium_float.onnx.zip
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 231587
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1e7cccec585914302ced86f86cff4ecae478415f5ec27d132f141173fb8fff58
|
| 3 |
size 231587
|
QuickSRNetMedium_w8a8.dlc
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:457177dac18f46eab6aa9092c3c02588d8a8f57ba44d92ca1c56055dc7aee96c
|
| 3 |
+
size 106220
|
README.md
CHANGED
|
@@ -36,62 +36,65 @@ 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 |
-
| QuickSRNetMedium | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 2.
|
| 40 |
-
| QuickSRNetMedium | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 2.
|
| 41 |
-
| QuickSRNetMedium | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 1.
|
| 42 |
-
| QuickSRNetMedium | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 1.
|
| 43 |
-
| QuickSRNetMedium | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 1.
|
| 44 |
-
| QuickSRNetMedium | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 0.
|
| 45 |
-
| QuickSRNetMedium | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 1.
|
| 46 |
-
| QuickSRNetMedium | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 5.
|
| 47 |
-
| QuickSRNetMedium | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC |
|
| 48 |
-
| QuickSRNetMedium | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 2.
|
| 49 |
-
| QuickSRNetMedium | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 2.
|
| 50 |
-
| QuickSRNetMedium | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 1.
|
| 51 |
-
| QuickSRNetMedium | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 0.
|
| 52 |
-
| QuickSRNetMedium | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 1.
|
| 53 |
-
| QuickSRNetMedium | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 1.
|
| 54 |
-
| QuickSRNetMedium | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 1.
|
| 55 |
-
| QuickSRNetMedium | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 0.
|
| 56 |
-
| QuickSRNetMedium | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 5.
|
| 57 |
-
| QuickSRNetMedium | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC |
|
| 58 |
-
| QuickSRNetMedium | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 0.
|
| 59 |
-
| QuickSRNetMedium | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 0.
|
| 60 |
-
| QuickSRNetMedium | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 0.776 ms | 0 -
|
| 61 |
-
| QuickSRNetMedium | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 0.
|
| 62 |
-
| QuickSRNetMedium | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 0.
|
| 63 |
-
| QuickSRNetMedium | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 0.
|
| 64 |
-
| QuickSRNetMedium | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 0.
|
| 65 |
-
| QuickSRNetMedium | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 0.
|
| 66 |
-
| QuickSRNetMedium | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 0.
|
| 67 |
-
| QuickSRNetMedium | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 0.
|
| 68 |
-
| QuickSRNetMedium | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 1.
|
| 69 |
-
| QuickSRNetMedium | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 0.
|
| 70 |
-
| QuickSRNetMedium | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 0.
|
| 71 |
-
| QuickSRNetMedium | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 0.
|
| 72 |
-
| QuickSRNetMedium | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 0.
|
| 73 |
-
| QuickSRNetMedium | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 0.
|
| 74 |
-
| QuickSRNetMedium | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 0.
|
| 75 |
-
| QuickSRNetMedium | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 0.
|
| 76 |
-
| QuickSRNetMedium | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 0.
|
| 77 |
-
| QuickSRNetMedium | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | TFLITE |
|
| 78 |
-
| QuickSRNetMedium | w8a8 |
|
| 79 |
-
| QuickSRNetMedium | w8a8 |
|
| 80 |
-
| QuickSRNetMedium | w8a8 | SA7255P ADP | Qualcomm® SA7255P |
|
| 81 |
-
| QuickSRNetMedium | w8a8 |
|
| 82 |
-
| QuickSRNetMedium | w8a8 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) |
|
| 83 |
-
| QuickSRNetMedium | w8a8 |
|
| 84 |
-
| QuickSRNetMedium | w8a8 | SA8295P ADP | Qualcomm® SA8295P |
|
| 85 |
-
| QuickSRNetMedium | w8a8 |
|
| 86 |
-
| QuickSRNetMedium | w8a8 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) |
|
| 87 |
-
| QuickSRNetMedium | w8a8 |
|
| 88 |
-
| QuickSRNetMedium | w8a8 | SA8775P ADP | Qualcomm® SA8775P |
|
| 89 |
-
| QuickSRNetMedium | w8a8 |
|
| 90 |
-
| QuickSRNetMedium | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile |
|
| 91 |
-
| QuickSRNetMedium | w8a8 | Samsung Galaxy
|
| 92 |
-
| QuickSRNetMedium | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile |
|
| 93 |
-
| QuickSRNetMedium | w8a8 |
|
| 94 |
-
| QuickSRNetMedium | w8a8 | Snapdragon
|
|
|
|
|
|
|
|
|
|
| 95 |
| QuickSRNetMedium | w8a8 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 0.447 ms | 6 - 6 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.dlc) |
|
| 96 |
|
| 97 |
|
|
@@ -106,9 +109,9 @@ pip install qai-hub-models
|
|
| 106 |
```
|
| 107 |
|
| 108 |
|
| 109 |
-
## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
|
| 110 |
|
| 111 |
-
Sign-in to [Qualcomm® AI Hub](https://
|
| 112 |
Qualcomm® ID. Once signed in navigate to `Account -> Settings -> API Token`.
|
| 113 |
|
| 114 |
With this API token, you can configure your client to run models on the cloud
|
|
@@ -116,7 +119,7 @@ hosted devices.
|
|
| 116 |
```bash
|
| 117 |
qai-hub configure --api_token API_TOKEN
|
| 118 |
```
|
| 119 |
-
Navigate to [docs](https://
|
| 120 |
|
| 121 |
|
| 122 |
|
|
@@ -227,7 +230,7 @@ With the output of the model, you can compute like PSNR, relative errors or
|
|
| 227 |
spot check the output with expected output.
|
| 228 |
|
| 229 |
**Note**: This on-device profiling and inference requires access to Qualcomm®
|
| 230 |
-
AI Hub. [Sign up for access](https://myaccount.qualcomm.com/signup).
|
| 231 |
|
| 232 |
|
| 233 |
|
|
|
|
| 36 |
|
| 37 |
| Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
|
| 38 |
|---|---|---|---|---|---|---|---|---|
|
| 39 |
+
| QuickSRNetMedium | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 2.983 ms | 0 - 16 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.tflite) |
|
| 40 |
+
| QuickSRNetMedium | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 2.484 ms | 0 - 17 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.dlc) |
|
| 41 |
+
| QuickSRNetMedium | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 1.286 ms | 0 - 26 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.tflite) |
|
| 42 |
+
| QuickSRNetMedium | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 1.148 ms | 0 - 26 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.dlc) |
|
| 43 |
+
| QuickSRNetMedium | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 1.057 ms | 0 - 5 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.tflite) |
|
| 44 |
+
| QuickSRNetMedium | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 0.803 ms | 0 - 6 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.dlc) |
|
| 45 |
+
| QuickSRNetMedium | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 1.213 ms | 0 - 8 MB | NPU | [QuickSRNetMedium.onnx.zip](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.onnx.zip) |
|
| 46 |
+
| QuickSRNetMedium | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 5.891 ms | 3 - 20 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.tflite) |
|
| 47 |
+
| QuickSRNetMedium | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 1.181 ms | 0 - 17 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.dlc) |
|
| 48 |
+
| QuickSRNetMedium | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 2.983 ms | 0 - 16 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.tflite) |
|
| 49 |
+
| QuickSRNetMedium | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 2.484 ms | 0 - 17 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.dlc) |
|
| 50 |
+
| QuickSRNetMedium | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 1.074 ms | 0 - 5 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.tflite) |
|
| 51 |
+
| QuickSRNetMedium | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 0.782 ms | 2 - 8 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.dlc) |
|
| 52 |
+
| QuickSRNetMedium | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 1.738 ms | 0 - 21 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.tflite) |
|
| 53 |
+
| QuickSRNetMedium | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 1.551 ms | 0 - 22 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.dlc) |
|
| 54 |
+
| QuickSRNetMedium | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 1.068 ms | 0 - 6 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.tflite) |
|
| 55 |
+
| QuickSRNetMedium | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 0.781 ms | 0 - 6 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.dlc) |
|
| 56 |
+
| QuickSRNetMedium | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 5.891 ms | 3 - 20 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.tflite) |
|
| 57 |
+
| QuickSRNetMedium | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 1.181 ms | 0 - 17 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.dlc) |
|
| 58 |
+
| QuickSRNetMedium | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 0.671 ms | 0 - 29 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.tflite) |
|
| 59 |
+
| QuickSRNetMedium | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 0.488 ms | 0 - 28 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.dlc) |
|
| 60 |
+
| QuickSRNetMedium | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 0.776 ms | 0 - 21 MB | NPU | [QuickSRNetMedium.onnx.zip](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.onnx.zip) |
|
| 61 |
+
| QuickSRNetMedium | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 0.486 ms | 0 - 23 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.tflite) |
|
| 62 |
+
| QuickSRNetMedium | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 0.379 ms | 0 - 21 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.dlc) |
|
| 63 |
+
| QuickSRNetMedium | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 0.617 ms | 0 - 17 MB | NPU | [QuickSRNetMedium.onnx.zip](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.onnx.zip) |
|
| 64 |
+
| QuickSRNetMedium | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 0.431 ms | 0 - 19 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.tflite) |
|
| 65 |
+
| QuickSRNetMedium | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 0.375 ms | 0 - 20 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.dlc) |
|
| 66 |
+
| QuickSRNetMedium | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 0.548 ms | 3 - 20 MB | NPU | [QuickSRNetMedium.onnx.zip](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.onnx.zip) |
|
| 67 |
+
| QuickSRNetMedium | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 0.9 ms | 1 - 1 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.dlc) |
|
| 68 |
+
| QuickSRNetMedium | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 1.108 ms | 8 - 8 MB | NPU | [QuickSRNetMedium.onnx.zip](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium.onnx.zip) |
|
| 69 |
+
| QuickSRNetMedium | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 0.959 ms | 1 - 17 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.tflite) |
|
| 70 |
+
| QuickSRNetMedium | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 0.856 ms | 0 - 16 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.dlc) |
|
| 71 |
+
| QuickSRNetMedium | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 0.516 ms | 1 - 26 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.tflite) |
|
| 72 |
+
| QuickSRNetMedium | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 0.521 ms | 0 - 24 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.dlc) |
|
| 73 |
+
| QuickSRNetMedium | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 0.466 ms | 0 - 4 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.tflite) |
|
| 74 |
+
| QuickSRNetMedium | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 0.354 ms | 0 - 11 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.dlc) |
|
| 75 |
+
| QuickSRNetMedium | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 0.652 ms | 0 - 17 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.tflite) |
|
| 76 |
+
| QuickSRNetMedium | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 0.534 ms | 0 - 17 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.dlc) |
|
| 77 |
+
| QuickSRNetMedium | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | TFLITE | 1.016 ms | 0 - 19 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.tflite) |
|
| 78 |
+
| QuickSRNetMedium | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | QNN_DLC | 1.012 ms | 0 - 21 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.dlc) |
|
| 79 |
+
| QuickSRNetMedium | w8a8 | RB5 (Proxy) | Qualcomm® QCS8250 (Proxy) | TFLITE | 13.372 ms | 1 - 4 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.tflite) |
|
| 80 |
+
| QuickSRNetMedium | w8a8 | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 0.959 ms | 1 - 17 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.tflite) |
|
| 81 |
+
| QuickSRNetMedium | w8a8 | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 0.856 ms | 0 - 16 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.dlc) |
|
| 82 |
+
| QuickSRNetMedium | w8a8 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 0.434 ms | 0 - 3 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.tflite) |
|
| 83 |
+
| QuickSRNetMedium | w8a8 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 0.364 ms | 0 - 4 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.dlc) |
|
| 84 |
+
| QuickSRNetMedium | w8a8 | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 0.833 ms | 0 - 21 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.tflite) |
|
| 85 |
+
| QuickSRNetMedium | w8a8 | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 0.746 ms | 0 - 22 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.dlc) |
|
| 86 |
+
| QuickSRNetMedium | w8a8 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 0.44 ms | 0 - 10 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.tflite) |
|
| 87 |
+
| QuickSRNetMedium | w8a8 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 0.34 ms | 0 - 3 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.dlc) |
|
| 88 |
+
| QuickSRNetMedium | w8a8 | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 0.652 ms | 0 - 17 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.tflite) |
|
| 89 |
+
| QuickSRNetMedium | w8a8 | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 0.534 ms | 0 - 17 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.dlc) |
|
| 90 |
+
| QuickSRNetMedium | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 0.276 ms | 0 - 27 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.tflite) |
|
| 91 |
+
| QuickSRNetMedium | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 0.22 ms | 0 - 27 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.dlc) |
|
| 92 |
+
| QuickSRNetMedium | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 0.232 ms | 0 - 20 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.tflite) |
|
| 93 |
+
| QuickSRNetMedium | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 0.172 ms | 0 - 24 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.dlc) |
|
| 94 |
+
| QuickSRNetMedium | w8a8 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | TFLITE | 0.464 ms | 0 - 22 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.tflite) |
|
| 95 |
+
| QuickSRNetMedium | w8a8 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | QNN_DLC | 0.37 ms | 0 - 23 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.dlc) |
|
| 96 |
+
| QuickSRNetMedium | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 0.173 ms | 0 - 19 MB | NPU | [QuickSRNetMedium.tflite](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.tflite) |
|
| 97 |
+
| QuickSRNetMedium | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 0.151 ms | 0 - 18 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.dlc) |
|
| 98 |
| QuickSRNetMedium | w8a8 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 0.447 ms | 6 - 6 MB | NPU | [QuickSRNetMedium.dlc](https://huggingface.co/qualcomm/QuickSRNetMedium/blob/main/QuickSRNetMedium_w8a8.dlc) |
|
| 99 |
|
| 100 |
|
|
|
|
| 109 |
```
|
| 110 |
|
| 111 |
|
| 112 |
+
## Configure Qualcomm® AI Hub Workbench to run this model on a cloud-hosted device
|
| 113 |
|
| 114 |
+
Sign-in to [Qualcomm® AI Hub Workbench](https://workbench.aihub.qualcomm.com/) with your
|
| 115 |
Qualcomm® ID. Once signed in navigate to `Account -> Settings -> API Token`.
|
| 116 |
|
| 117 |
With this API token, you can configure your client to run models on the cloud
|
|
|
|
| 119 |
```bash
|
| 120 |
qai-hub configure --api_token API_TOKEN
|
| 121 |
```
|
| 122 |
+
Navigate to [docs](https://workbench.aihub.qualcomm.com/docs/) for more information.
|
| 123 |
|
| 124 |
|
| 125 |
|
|
|
|
| 230 |
spot check the output with expected output.
|
| 231 |
|
| 232 |
**Note**: This on-device profiling and inference requires access to Qualcomm®
|
| 233 |
+
AI Hub Workbench. [Sign up for access](https://myaccount.qualcomm.com/signup).
|
| 234 |
|
| 235 |
|
| 236 |
|
tool-versions.yaml
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
tool_versions:
|
| 2 |
qnn_dlc:
|
| 3 |
-
qairt: 2.
|
|
|
|
| 1 |
tool_versions:
|
| 2 |
qnn_dlc:
|
| 3 |
+
qairt: 2.40.0.251030114326_189385
|