File size: 763 Bytes
df2a0ea ba037b4 928bfc4 ba037b4 928bfc4 ba037b4 928bfc4 510ad92 ba037b4 510ad92 ba037b4 928bfc4 ba037b4 928bfc4 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 |
---
license: apache-2.0
---
### Imports
```python
import torch
from hqq.engine.hf import HQQModelForCausalLM, AutoTokenizer
from hqq.utils.patching import prepare_for_inference
```
### Loading Weights
```python
model = HQQModelForCausalLM.from_quantized("rohitg/Mixtral-8x22B-Instruct-v0.1-hf-4bit_g64-HQQ", device='cuda')
tokenizer = AutoTokenizer.from_pretrained('mistralai/Mixtral-8x22B-Instruct-v0.1')
prepare_for_inference(model, backend="torchao_int4")
```
### Text Generation
```python
prompt = "<s> [INST] How do I build a car? [/INST] "
inputs = tokenizer(prompt, return_tensors="pt", add_special_tokens=False)
outputs = model.generate(**(inputs.to('cuda')), max_new_tokens=1000)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
|