49 lines
1.1 KiB
Markdown
49 lines
1.1 KiB
Markdown
---
|
||
tags:
|
||
- gptq
|
||
- 4bit
|
||
- int4
|
||
- gptqmodel
|
||
- modelcloud
|
||
- llama-3.1
|
||
- 8b
|
||
- base
|
||
license: llama3.1
|
||
---
|
||
This model has been quantized using [GPTQModel](https://github.com/ModelCloud/GPTQModel).
|
||
|
||
- **bits**: 4
|
||
- **group_size**: 128
|
||
- **desc_act**: true
|
||
- **static_groups**: false
|
||
- **sym**: true
|
||
- **lm_head**: false
|
||
- **damp_percent**: 0.01
|
||
- **true_sequential**: true
|
||
- **model_name_or_path**: ""
|
||
- **model_file_base_name**: "model"
|
||
- **quant_method**: "gptq"
|
||
- **checkpoint_format**: "gptq"
|
||
- **meta**：
|
||
- **quantizer**: "gptqmodel:0.9.9-dev0"
|
||
|
||
**Here is an example:**
|
||
```python
|
||
import torch
|
||
from transformers import AutoTokenizer
|
||
from gptqmodel import GPTQModel
|
||
|
||
device = torch.device("cuda:0")
|
||
|
||
model_name = "ModelCloud/Meta-Llama-3.1-8B-gptq-4bit"
|
||
|
||
prompt = "I am in Shanghai, preparing to visit the natural history museum. Can you tell me the best way to"
|
||
|
||
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
||
|
||
model = GPTQModel.from_quantized(model_name)
|
||
|
||
inputs = tokenizer(prompt, return_tensors="pt").to(device)
|
||
res = model.generate(**inputs, num_beams=1, min_new_tokens=1, max_new_tokens=512)
|
||
print(tokenizer.decode(res[0]))
|
||
``` |