Buy Question n Answering Demo using BERT in Python + Flask

If you are looking to setup same demo like our NLP Question And Answering then we can provide you with the code, fine-tuned model and all required setup instructions with nominal charges.

BERT based NLP QnA Demo

English Version

  • Source Code of Python + Flask (Tensorflow based)
  • BERT Model Fine Tuned on SQuAD 2.0 English Dataset
  • Easy to setup code

USD $99 Only

Other language

  • Source Code of Python + Flask (Tensorflow based)
  • BERT Model Fine Tuned on thousands of questions for the language you purchase
  • Languages: Hindi, Arabic, Spanish, French, German, Chinese, Portuguese
  • Easy to setup code

USD $199 Only / Language


  • If you want to setup the demo on your own then you can follow our NLP Tutorial: Setup Question Answering System using BERT + SQuAD on Colab TPU which will take you through how to fine-tune the BERT model on SQuAD 2.0 and how to do the inference on your own paragraph and questions. The tutorial doesn't include instruction about creating web interface for QnA like our demo. Our package will provide the Python + Flask code which is not included in our tutorial.
  • Please test the result thoroughly on the demo for the language you are planning to purchase.
  • The demo contains API URL using which you can call it from your external application/interface.
  • The answers may slightly vary from the demo as the fine-tuning of the models were done with different computational resources.
  • You agree to not to resale the code and fine-tuned model or distribute them freely. It is for personal usage only.
  • Performance of this model is not production ready. It takes long to generate result. So it's suitable for learning purpose only.

The below table shows how the prediction time changes with the BERT English model and BERT Multilingual Model on different configuration systems.

4 Cores CPU & 15 GB RAM

(Google Cloud)

2 Cores CPU & 7.5 GB RAM

(Google Cloud)

8th Generation i3 processor & 8GB RAM

(Local System)

Azure NC6 – 1K80 GPU


25 Seconds

45 Seconds

35 Seconds

10 Seconds


17 Seconds

32 seconds

20 Seconds

7 Seconds

NOTE: Prediction time can also vary by tuning the Hyper-parameters like Batch-Size or Max_seq_length but you need to be careful to make changes in hyper-parameters as it can also affect the accuracy.

  • The QnA package DOESN'T INCLUDE DATASETS of Hindi, Arabic, German, Spanish, French or Chinese language.

Interested to purchase? please email: