We created FlexCheck AI to provide scoring and feedback for Writing and Speaking. We designed it to follow the IELTS marking criteria, like an examiner would, and we trained it using our huge database of real student answers.
Our AI is more accurate than free online LLMs, such as Chat GPT or Deepseek. It is also more consistent. Across multiple submissions using FlexCheck AI, we believe you will have an accurate idea of your level.
However, our AI is not as accurate as a human marker, yet! If you would like feedback from one of our IELTS expert tutors, you can get this on IELTS FLEX.
Sometimes, the results of your practice or mock test might be a little too high or a little too low. Like humans, the AI can make errors, and so the best way to be sure of your level is to take more than one mock test / do more than one practice.
Like a scientist, you cannot be sure of your result from only one test!
I disagree with my score!
Occasionally, students disagree with their score from AI. This may be that the AI has made a small error, or it may be that the score is correct, but not what the student wanted! The best thing to do is to do another practice. If you believe you are regularly getting the wrong score, please message us. Use the Chatbot and, if possible, include the URL (web address) of your practice, the date you took it, and the type of task you did - we'd be happy to investigate.
I tested it with a model answer and it gave the wrong score!
We used some of our model answers within the training of the bot, so there is a chance that you are getting an incorrect score due to the AI detecting copied answers, or irrelevant answers. Our AI performs best with real student responses, not copied ones.
I don't understand the feedback!
If the Advantage feedback comments contain information which is confusing, or seems wrong, let us know! It is very unusual, but sometimes the AI (like all AI) says things which don't quite fit. Feel free to ask us about this. Use the Chatbot to send us a message and, if possible, include the URL (web address) of your practice, the date you took it, and the type of task you did - we'd be happy to investigate.
The AI never gives high band scores!
Writing is hard. The average global score in 2023-2024 for Writing is just under 6.0, and for Speaking it is just over 6.0.
Achieving a 7.0 or higher requires a good understanding of the marking criteria, and a C1+ level of English.
As an example, during 31 days of January 2025:
- 0.02% of student Writing answers scored 9.0 with FlexCheck AI
- 5.2% of student Writing answers scored 8.0 with FlexCheck AI
- 16.4% of student Writing answers scored 7.0 with FlexCheck AI
- 0.18% of student Speaking answers scored 9.0 with FlexCheck AI
- 6.1% of student Speaking answers scored 8.0 with FlexCheck AI
- 11.5% of student Speaking answers scored 7.0 with FlexCheck AI
Keep practising, and we hope FlexCheck AI is a useful tool to help you reach your target!
Was this article helpful?
That’s Great!
Thank you for your feedback
Sorry! We couldn't be helpful
Thank you for your feedback
Feedback sent
We appreciate your effort and will try to fix the article