metadata
license: mit
tags:
- LiteRT
base_model:
- deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
litert-community/DeepSeek-R1-Distill-Qwen-1.5B
This model was converted to LiteRT (aka TFLite) format from deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B using Google AI Edge Torch.
Run the model in colab
Disclaimer: Model performance demonstrated with the Python API in this notebook is not representative of performance on a local device.
Run the model on Android
Please follow the instructions.
Benchmarking results
Note that all benchmark stats are from a Samsung S24 Ultra.
| Model | DeepSeek-R1-Distill-Qwen-1.5B (Int8 quantized) | |
|---|---|---|
| Params | 1.78 B | |
| Prefill 512 tokens | Decode 128 tokens | |
| LiteRT tk/s (XNNPACK, 4 threads) | 260.95 | 23.126 |
| GGML tk/s (CPU, 4 threads) | 64.66 | 23.85 |