Chattruth 7B

🚀 ChatTruth-7B

ChatTruth-7B is optimized and trained on the basis of Qwen-VL with carefully designed data. Compared with Qwen-VL, the model has been significantly improved in high-resolution scenarios. It innovatively proposes the Restore Module, which greatly reduces the computational cost for high-resolution processing.

🚀 Quick Start

📦 Installation

💻 Usage Examples

Basic Usage

from transformers import AutoModelForCausalLM, AutoTokenizer
from transformers.generation import GenerationConfig
import torch
torch.manual_seed(1234)
model_path = 'ChatTruth-7B' # your downloaded model path.

tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)

# use cuda device
model = AutoModelForCausalLM.from_pretrained(model_path, device_map="cuda", trust_remote_code=True).eval()

model.generation_config = GenerationConfig.from_pretrained(model_path, trust_remote_code=True)
model.generation_config.top_p = 0.01

query = tokenizer.from_list_format([
    {'image': 'demo.jpeg'},
    {'text': '图片中的文字是什么'},
])
response, history = model.chat(tokenizer, query=query, history=None)
print(response)

# 昆明太厉害了