# Train a model for document specific Q and A

**URL:** <https://discuss.huggingface.co/t/train-a-model-for-document-specific-q-and-a/32015>\
**Category:** Community Calls\
**Created:** [February 19, 2023, 8:57am UTC](https://discuss.huggingface.co/t/train-a-model-for-document-specific-q-and-a/32015 "2023-02-19T08:57:57Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![dato3000](https://avatars.discourse-cdn.com/v4/letter/d/b5e925/32.png) [@dato3000](https://discuss.huggingface.co/u/dato3000)\
**Post date:** [February 19, 2023, 8:57am UTC](https://discuss.huggingface.co/t/train-a-model-for-document-specific-q-and-a/32015/1 "2023-02-19T08:57:57Z")

</div>

is there a way to train/fine tune a LLM model so that it has built into it a specific context (i.e. a book), and then when you ask the GPT questions, it already knows about the novel, and answers specifically to ONLY that novel? (this way you don’t need to include the entire book in the prompt every time you ask a question)