v0.42.0
Browse filesSee https://github.com/quic/ai-hub-models/releases/v0.42.0 for changelog.
- Nomic-Embed-Text_float.dlc +2 -2
- Nomic-Embed-Text_float.onnx.zip +2 -2
- README.md +34 -34
Nomic-Embed-Text_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:cba0791a7ed174ab45581bcd76c6b03ef7a730b5f49585da844d39c41e84603f
|
| 3 |
+
size 548667644
|
Nomic-Embed-Text_float.onnx.zip
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:b0225fae85323fe1fd27a9de9334c23d7f54d1f0322696885508558c667e58d4
|
| 3 |
+
size 507828789
|
README.md
CHANGED
|
@@ -35,36 +35,36 @@ More details on model performance across various devices, can be found
|
|
| 35 |
|
| 36 |
| Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
|
| 37 |
|---|---|---|---|---|---|---|---|---|
|
| 38 |
-
| Nomic-Embed-Text | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 31.
|
| 39 |
-
| Nomic-Embed-Text | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 28.
|
| 40 |
-
| Nomic-Embed-Text | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 10.
|
| 41 |
-
| Nomic-Embed-Text | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 10.
|
| 42 |
-
| Nomic-Embed-Text | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE |
|
| 43 |
-
| Nomic-Embed-Text | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 7.
|
| 44 |
-
| Nomic-Embed-Text | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 8.
|
| 45 |
-
| Nomic-Embed-Text | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 11.
|
| 46 |
-
| Nomic-Embed-Text | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 9.
|
| 47 |
-
| Nomic-Embed-Text | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 31.
|
| 48 |
-
| Nomic-Embed-Text | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 28.
|
| 49 |
-
| Nomic-Embed-Text | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE |
|
| 50 |
-
| Nomic-Embed-Text | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 7.
|
| 51 |
-
| Nomic-Embed-Text | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 12.
|
| 52 |
-
| Nomic-Embed-Text | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 10.
|
| 53 |
-
| Nomic-Embed-Text | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 9.
|
| 54 |
-
| Nomic-Embed-Text | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 7.
|
| 55 |
-
| Nomic-Embed-Text | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 11.
|
| 56 |
-
| Nomic-Embed-Text | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 9.
|
| 57 |
-
| Nomic-Embed-Text | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 6.
|
| 58 |
-
| Nomic-Embed-Text | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 5.
|
| 59 |
-
| Nomic-Embed-Text | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX |
|
| 60 |
-
| Nomic-Embed-Text | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 5.
|
| 61 |
-
| Nomic-Embed-Text | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 4.
|
| 62 |
-
| Nomic-Embed-Text | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 4.
|
| 63 |
-
| Nomic-Embed-Text | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 4.
|
| 64 |
-
| Nomic-Embed-Text | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 3.
|
| 65 |
-
| Nomic-Embed-Text | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 4.
|
| 66 |
-
| Nomic-Embed-Text | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC |
|
| 67 |
-
| Nomic-Embed-Text | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX |
|
| 68 |
|
| 69 |
|
| 70 |
|
|
@@ -79,9 +79,9 @@ pip install "qai-hub-models[nomic-embed-text]"
|
|
| 79 |
```
|
| 80 |
|
| 81 |
|
| 82 |
-
## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
|
| 83 |
|
| 84 |
-
Sign-in to [Qualcomm® AI Hub](https://
|
| 85 |
Qualcomm® ID. Once signed in navigate to `Account -> Settings -> API Token`.
|
| 86 |
|
| 87 |
With this API token, you can configure your client to run models on the cloud
|
|
@@ -89,7 +89,7 @@ hosted devices.
|
|
| 89 |
```bash
|
| 90 |
qai-hub configure --api_token API_TOKEN
|
| 91 |
```
|
| 92 |
-
Navigate to [docs](https://
|
| 93 |
|
| 94 |
|
| 95 |
|
|
@@ -200,7 +200,7 @@ With the output of the model, you can compute like PSNR, relative errors or
|
|
| 200 |
spot check the output with expected output.
|
| 201 |
|
| 202 |
**Note**: This on-device profiling and inference requires access to Qualcomm®
|
| 203 |
-
AI Hub. [Sign up for access](https://myaccount.qualcomm.com/signup).
|
| 204 |
|
| 205 |
|
| 206 |
|
|
|
|
| 35 |
|
| 36 |
| Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
|
| 37 |
|---|---|---|---|---|---|---|---|---|
|
| 38 |
+
| Nomic-Embed-Text | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 31.85 ms | 0 - 366 MB | NPU | [Nomic-Embed-Text.tflite](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.tflite) |
|
| 39 |
+
| Nomic-Embed-Text | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 28.278 ms | 0 - 363 MB | NPU | [Nomic-Embed-Text.dlc](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.dlc) |
|
| 40 |
+
| Nomic-Embed-Text | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 10.943 ms | 0 - 373 MB | NPU | [Nomic-Embed-Text.tflite](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.tflite) |
|
| 41 |
+
| Nomic-Embed-Text | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 10.899 ms | 0 - 369 MB | NPU | [Nomic-Embed-Text.dlc](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.dlc) |
|
| 42 |
+
| Nomic-Embed-Text | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 8.752 ms | 0 - 26 MB | NPU | [Nomic-Embed-Text.tflite](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.tflite) |
|
| 43 |
+
| Nomic-Embed-Text | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 7.54 ms | 0 - 27 MB | NPU | [Nomic-Embed-Text.dlc](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.dlc) |
|
| 44 |
+
| Nomic-Embed-Text | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 8.432 ms | 0 - 661 MB | NPU | [Nomic-Embed-Text.onnx.zip](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.onnx.zip) |
|
| 45 |
+
| Nomic-Embed-Text | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 11.214 ms | 0 - 365 MB | NPU | [Nomic-Embed-Text.tflite](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.tflite) |
|
| 46 |
+
| Nomic-Embed-Text | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 9.89 ms | 0 - 365 MB | NPU | [Nomic-Embed-Text.dlc](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.dlc) |
|
| 47 |
+
| Nomic-Embed-Text | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 31.85 ms | 0 - 366 MB | NPU | [Nomic-Embed-Text.tflite](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.tflite) |
|
| 48 |
+
| Nomic-Embed-Text | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 28.278 ms | 0 - 363 MB | NPU | [Nomic-Embed-Text.dlc](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.dlc) |
|
| 49 |
+
| Nomic-Embed-Text | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 9.022 ms | 0 - 15 MB | NPU | [Nomic-Embed-Text.tflite](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.tflite) |
|
| 50 |
+
| Nomic-Embed-Text | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 7.487 ms | 0 - 28 MB | NPU | [Nomic-Embed-Text.dlc](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.dlc) |
|
| 51 |
+
| Nomic-Embed-Text | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 12.485 ms | 0 - 358 MB | NPU | [Nomic-Embed-Text.tflite](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.tflite) |
|
| 52 |
+
| Nomic-Embed-Text | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 10.698 ms | 0 - 359 MB | NPU | [Nomic-Embed-Text.dlc](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.dlc) |
|
| 53 |
+
| Nomic-Embed-Text | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 9.046 ms | 0 - 41 MB | NPU | [Nomic-Embed-Text.tflite](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.tflite) |
|
| 54 |
+
| Nomic-Embed-Text | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 7.533 ms | 0 - 25 MB | NPU | [Nomic-Embed-Text.dlc](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.dlc) |
|
| 55 |
+
| Nomic-Embed-Text | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 11.214 ms | 0 - 365 MB | NPU | [Nomic-Embed-Text.tflite](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.tflite) |
|
| 56 |
+
| Nomic-Embed-Text | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 9.89 ms | 0 - 365 MB | NPU | [Nomic-Embed-Text.dlc](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.dlc) |
|
| 57 |
+
| Nomic-Embed-Text | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 6.433 ms | 0 - 373 MB | NPU | [Nomic-Embed-Text.tflite](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.tflite) |
|
| 58 |
+
| Nomic-Embed-Text | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 5.26 ms | 0 - 371 MB | NPU | [Nomic-Embed-Text.dlc](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.dlc) |
|
| 59 |
+
| Nomic-Embed-Text | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 6.004 ms | 0 - 373 MB | NPU | [Nomic-Embed-Text.onnx.zip](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.onnx.zip) |
|
| 60 |
+
| Nomic-Embed-Text | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 5.404 ms | 0 - 367 MB | NPU | [Nomic-Embed-Text.tflite](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.tflite) |
|
| 61 |
+
| Nomic-Embed-Text | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 4.208 ms | 0 - 366 MB | NPU | [Nomic-Embed-Text.dlc](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.dlc) |
|
| 62 |
+
| Nomic-Embed-Text | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 4.741 ms | 0 - 368 MB | NPU | [Nomic-Embed-Text.onnx.zip](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.onnx.zip) |
|
| 63 |
+
| Nomic-Embed-Text | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 4.773 ms | 0 - 364 MB | NPU | [Nomic-Embed-Text.tflite](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.tflite) |
|
| 64 |
+
| Nomic-Embed-Text | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 3.865 ms | 0 - 364 MB | NPU | [Nomic-Embed-Text.dlc](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.dlc) |
|
| 65 |
+
| Nomic-Embed-Text | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 4.201 ms | 0 - 364 MB | NPU | [Nomic-Embed-Text.onnx.zip](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.onnx.zip) |
|
| 66 |
+
| Nomic-Embed-Text | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 8.088 ms | 1491 - 1491 MB | NPU | [Nomic-Embed-Text.dlc](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.dlc) |
|
| 67 |
+
| Nomic-Embed-Text | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 8.993 ms | 263 - 263 MB | NPU | [Nomic-Embed-Text.onnx.zip](https://huggingface.co/qualcomm/Nomic-Embed-Text/blob/main/Nomic-Embed-Text.onnx.zip) |
|
| 68 |
|
| 69 |
|
| 70 |
|
|
|
|
| 79 |
```
|
| 80 |
|
| 81 |
|
| 82 |
+
## Configure Qualcomm® AI Hub Workbench to run this model on a cloud-hosted device
|
| 83 |
|
| 84 |
+
Sign-in to [Qualcomm® AI Hub Workbench](https://workbench.aihub.qualcomm.com/) with your
|
| 85 |
Qualcomm® ID. Once signed in navigate to `Account -> Settings -> API Token`.
|
| 86 |
|
| 87 |
With this API token, you can configure your client to run models on the cloud
|
|
|
|
| 89 |
```bash
|
| 90 |
qai-hub configure --api_token API_TOKEN
|
| 91 |
```
|
| 92 |
+
Navigate to [docs](https://workbench.aihub.qualcomm.com/docs/) for more information.
|
| 93 |
|
| 94 |
|
| 95 |
|
|
|
|
| 200 |
spot check the output with expected output.
|
| 201 |
|
| 202 |
**Note**: This on-device profiling and inference requires access to Qualcomm®
|
| 203 |
+
AI Hub Workbench. [Sign up for access](https://myaccount.qualcomm.com/signup).
|
| 204 |
|
| 205 |
|
| 206 |
|