tags: | |
- deepsparse | |
# mpt_7b_chat-dense_quant_linearW8A8MatMul8Embeds8LMhead8 | |
```python | |
import deepsparse | |
from huggingface_hub import snapshot_download | |
MODEL_PATH = snapshot_download(repo_id="mgoin/mpt-7b-chat-quant") | |
model = deepsparse.Pipeline.create(task="text-generation", model_path=MODEL_PATH) | |
model(sequences="Tell me a joke.") | |
``` |