# Chat with PDF locally using Llama 3

**URL:** <https://discuss.huggingface.co/t/chat-with-pdf-locally-using-llama-3/85218>\
**Category:** Show and Tell\
**Created:** [May 5, 2024, 5:42am UTC](https://discuss.huggingface.co/t/chat-with-pdf-locally-using-llama-3/85218 "2024-05-05T05:42:06Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![xiaoyich](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/xiaoyich/32/26603_2.png) [@xiaoyich](https://discuss.huggingface.co/u/xiaoyich)\
**Post date:** [May 5, 2024, 5:42am UTC](https://discuss.huggingface.co/t/chat-with-pdf-locally-using-llama-3/85218/1 "2024-05-05T05:42:06Z")

</div>

Hi everyone,

Recently, we added chat with PDF feature, local RAG and Llama 3 support in [RecurseChat](https://recurse.chat/), a local AI chat app on macOS.

I wrote about why we build it and the technical details here: [Local Docs, Local AI: Chat with PDF locally using Llama 3](https://medium.com/@recursechat/local-docs-local-ai-chat-with-pdf-locally-using-llama-3-a3b9a3a0c055).

You can chat with PDF locally and offline with built-in models such as Meta Llama 3 and Mistral, your own GGUF models or online providers like Together AI and Groq.

Looking forward to your feedback!