client.py

import requests
from generate_jwt import create_access_token

token = create_access_token({'sub': 'colab_user'})
headers = {'Authorization': f'Bearer {token}'}

data = {
    'messages': [{'role': 'user', 'content': 'Explain transformers in one sentence.'}],
    'enable_thinking': False
}

try:
    response = requests.post('http://127.0.0.1:8000/generate', json=data, headers=headers)
    print(f'Status: {response.status_code}')
    print(response.json())
except Exception as e:
    print(f'Error: {e}')