Views
No views yet
Pillow==10.1.0
torch==2.1.2
torchvision==0.16.2
transformers==4.40.0
sentencepiece==0.1.99
accelerate==0.30.1
bitsandbytes==0.43.11# test.py
2import torch
3from PIL import Image
4from transformers import AutoModel, AutoTokenizer
5
6model = AutoModel.from_pretrained('openbmb/MiniCPM-Llama3-V-2_5-int4', trust_remote_code=True)
7tokenizer = AutoTokenizer.from_pretrained('openbmb/MiniCPM-Llama3-V-2_5-int4', trust_remote_code=True)
8model.eval()
9
10image = Image.open('xx.jpg').convert('RGB')
11question = 'What is in the image?'
12msgs = [{'role': 'user', 'content': question}]
13
14res = model.chat(
15 image=image,
16 msgs=msgs,
17 tokenizer=tokenizer,
18 sampling=True, # if sampling=False, beam_search will be used by default
19 temperature=0.7,
20 # system_prompt='' # pass system_prompt if needed
21)
22print(res)
23
24## if you want to use streaming, please make sure sampling=True and stream=True
25## the model.chat will return a generator
26res = model.chat(
27 image=image,
28 msgs=msgs,
29 tokenizer=tokenizer,
30 sampling=True,
31 temperature=0.7,
32 stream=True
33)
34
35generated_text = ""
36for new_text in res:
37 generated_text += new_text
38 print(new_text, flush=True, end='')