Huggingface Serverless free LLM inference API

from flask import Flask, request, jsonify, send_file
import requests

app = Flask(__name__)

LLModel = "bigscience/bloom"
ApiUrl = "https://api-inference.huggingface.co/models/" + LLModel
minOutputTokens = 500000
headers = {"Authorization": "Bearer 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"}


past_user_inputs = []
generated_responses = []


def query(api_url, headers, payload, min_length):
    try:
        payload["min_length"] = min_length
        response = requests.post(api_url, headers=headers, json=payload)
        response.raise_for_status() 
        return response.json()
    except requests.exceptions.RequestException as e:
        print("Error in processing the request:", e)
        return None


@app.route('/')
def index():
    
    return send_file('static/index.html')


@app.route('/chat', methods=['POST'])
def chat():
    user_message = request.json.get('message')
    model_path = ApiUrl

    if model_path:
        
        payload = {
            "inputs": user_message,
            "past_user_inputs": past_user_inputs,
            "generated_responses": generated_responses
        }

        bot_response = query(model_path, headers=headers, payload=payload, min_length=minOutputTokens)
        if bot_response is not None:
            past_user_inputs.append(user_message)
            generated_responses.append(bot_response[0]['generated_text'])

            
            response_json = {"message": bot_response,
                            "past_user_inputs": past_user_inputs,
                            "generated_responses": generated_responses}
            print(response_json)
            return jsonify(response_json)
        else:
            return jsonify({"message": "Error in processing the request."}), 500
    else:
        return jsonify({"message": "Model path not provided."}), 400


if __name__ == '__main__':
    app.run(debug=True)

python

I really cant figure out why my min_length doesnt at all effect the output...
thanks