proxy_model.py

import os
import requests
import torch
from transformers import AutoTokenizer, AutoConfig, AutoModelForCausalLM

class RemoteModelProxy:
    def __init__(self, model_id):
        self.model_id = model_id
        self.tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)
        
        # Load the configuration and remove the quantization configuration
        config = AutoConfig.from_pretrained(model_id, trust_remote_code=True)
        if hasattr(config, 'quantization_config'):
            del config.quantization_config
        
        self.config = config
        self.model = AutoModelForCausalLM.from_pretrained(model_id, config=self.config, trust_remote_code=True)

    def classify_text(self, text):
        inputs = self.tokenizer(text, return_tensors="pt", padding=True, truncation=True)
        logits = self.model(**inputs)
        probabilities = torch.softmax(logits, dim=-1).tolist()[0]
        predicted_class = torch.argmax(logits, dim=-1).item()
        return {
            "Predicted Class": predicted_class,
            "Probabilities": probabilities
        }

if __name__ == "__main__":
    model_id = "deepseek-ai/DeepSeek-V3"
    proxy = RemoteModelProxy(model_id)
    result = proxy.classify_text("Your input text here")
    print(result)