Nomic-Embed-Text: Optimized for Qualcomm Devices
A text encoder that surpasses OpenAI text-embedding-ada-002 and text-embedding-3-small performance on short and long context tasks.
This is based on the implementation of Nomic-Embed-Text found here. This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the Qualcomm® AI Hub Models library to export with custom configurations. More details on model performance across various devices, can be found here.
Qualcomm AI Hub Models uses Qualcomm AI Hub Workbench to compile, profile, and evaluate this model. Sign up to run these models on a hosted Qualcomm® device.
Getting Started
There are two ways to deploy this model on your device:
Option 1: Download Pre-Exported Models
Below are pre-exported model assets ready for deployment.
| Runtime | Precision | Chipset | SDK Versions | Download |
|---|---|---|---|---|
| ONNX | float | Universal | QAIRT 2.42, ONNX Runtime 1.24.3 | Download |
| QNN_DLC | float | Universal | QAIRT 2.45 | Download |
| TFLITE | float | Universal | QAIRT 2.45 | Download |
For more device-specific assets and performance metrics, visit Nomic-Embed-Text on Qualcomm® AI Hub.
Option 2: Export with Custom Configurations
Use the Qualcomm® AI Hub Models Python library to compile and export the model with your own:
- Custom weights (e.g., fine-tuned checkpoints)
- Custom input shapes
- Target device and runtime configurations
This option is ideal if you need to customize the model beyond the default configuration provided here.
See our repository for Nomic-Embed-Text on GitHub for usage instructions.
Model Details
Model Type: Model_use_case.text_generation
Model Stats:
- Model checkpoint: v1.5
- Input resolution: 1x128 (seqlen can vary)
- Number of parameters: 137M
- Model size (float): 523 MB
Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |
|---|---|---|---|---|---|---|
| Nomic-Embed-Text | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.86 ms | 0 - 416 MB | NPU |
| Nomic-Embed-Text | ONNX | float | Snapdragon® X2 Elite | 3.573 ms | 263 - 263 MB | NPU |
| Nomic-Embed-Text | ONNX | float | Snapdragon® X Elite | 8.506 ms | 263 - 263 MB | NPU |
| Nomic-Embed-Text | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 5.573 ms | 0 - 462 MB | NPU |
| Nomic-Embed-Text | ONNX | float | Qualcomm® QCS8550 (Proxy) | 7.973 ms | 0 - 324 MB | NPU |
| Nomic-Embed-Text | ONNX | float | Qualcomm® QCS9075 | 10.587 ms | 0 - 3 MB | NPU |
| Nomic-Embed-Text | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 4.338 ms | 0 - 414 MB | NPU |
| Nomic-Embed-Text | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.946 ms | 0 - 184 MB | NPU |
| Nomic-Embed-Text | QNN_DLC | float | Snapdragon® X2 Elite | 3.799 ms | 0 - 0 MB | NPU |
| Nomic-Embed-Text | QNN_DLC | float | Snapdragon® X Elite | 8.025 ms | 0 - 0 MB | NPU |
| Nomic-Embed-Text | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 5.279 ms | 0 - 373 MB | NPU |
| Nomic-Embed-Text | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 28.297 ms | 0 - 178 MB | NPU |
| Nomic-Embed-Text | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 7.412 ms | 0 - 2 MB | NPU |
| Nomic-Embed-Text | QNN_DLC | float | Qualcomm® SA8775P | 9.631 ms | 0 - 177 MB | NPU |
| Nomic-Embed-Text | QNN_DLC | float | Qualcomm® QCS9075 | 10.365 ms | 0 - 2 MB | NPU |
| Nomic-Embed-Text | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 10.957 ms | 0 - 351 MB | NPU |
| Nomic-Embed-Text | QNN_DLC | float | Qualcomm® SA7255P | 28.297 ms | 0 - 178 MB | NPU |
| Nomic-Embed-Text | QNN_DLC | float | Qualcomm® SA8295P | 10.597 ms | 0 - 155 MB | NPU |
| Nomic-Embed-Text | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 4.302 ms | 0 - 177 MB | NPU |
| Nomic-Embed-Text | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.948 ms | 0 - 190 MB | NPU |
| Nomic-Embed-Text | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 5.273 ms | 0 - 376 MB | NPU |
| Nomic-Embed-Text | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 28.287 ms | 0 - 183 MB | NPU |
| Nomic-Embed-Text | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 7.291 ms | 0 - 2 MB | NPU |
| Nomic-Embed-Text | TFLITE | float | Qualcomm® SA8775P | 9.684 ms | 0 - 183 MB | NPU |
| Nomic-Embed-Text | TFLITE | float | Qualcomm® QCS9075 | 10.569 ms | 0 - 265 MB | NPU |
| Nomic-Embed-Text | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 10.949 ms | 0 - 354 MB | NPU |
| Nomic-Embed-Text | TFLITE | float | Qualcomm® SA7255P | 28.287 ms | 0 - 183 MB | NPU |
| Nomic-Embed-Text | TFLITE | float | Qualcomm® SA8295P | 10.636 ms | 0 - 157 MB | NPU |
| Nomic-Embed-Text | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 4.317 ms | 0 - 186 MB | NPU |
License
- The license for the original implementation of Nomic-Embed-Text can be found here.
References
Community
- Join our AI Hub Slack community to collaborate, post questions and learn more about on-device AI.
- For questions or feedback please reach out to us.