How to Install PrivateGPT - AI Chat with your own Files! (Quick Setup and How to Use - EASY!)

PrivateGPT is a new trending GitHub project allowing you to use AI to Chat with your own Documents, on your own PC without Internet access. This AI GPT LLM runs of your CPU so you don't even need an expensive graphics card to play with PrivateGPT on your own PC. This is Opensource and 100% Free to use!
Ingest and interrogate the documents you work with daily and put PrivateGPT through its paces. Follow this Video for a Step by Step PrivateGPT Windows Install the Easy Way!

Links used in the Video:
GitHub PrivateGPT Link.
Git for Windows download.
Python download link:
Anaconda download link:
Visual Studio Community download link:

LLM Models you can use with Private GPT.
GPT4ALL - Groovy.
Vicuna 13B.

Commands used in Video:
Setup Private Conda Environment.
python --version
conda create -n privategpt python=3.11.4
conda activate privategpt

Clone and Install PrivateGPT.

Ingest Documents and Run PrivateGPT.
python privateGPT.py

[stuffaboutstuff]

Hi Everyone,
I uploaded a new PrivateGPT v 0.4.0 build Video, This Video shows how to build PrivateGPT on Windows with Ollama support. Check it out on the link below.
🔗PrivateGPT 0.4.0 with Ollama Support.

[machinimaaquinix]

Very well done and concise. Gonna give it a try today! And it all worked as of Sept 14, 2023. (its crazy how fast AI tutes on YT break given how fast things are changing.) My old 8th gen intel i3 is struggling with this but when given a few min to think, the results are quite good.

[Jamesbond-ohig]

Hello I running this issue "pip3 install -r requirements.txt
ERROR: Could not open requirements file: [Errno 2] No such file or directory: 'requirements.txt'"

[stuffaboutstuff]

Hi Everyone,
Please find a link to updated build video for the New version of PrivateGPT (Jan2024 for Windows).
If you want to build the original version you can follow the links in the video posted above and look for the Primordial version on the GitHub page.

[doctorywhy]

Thanks for the work. I tested it but got some issues. If I set MODEL_TYPE to LlamaC, I got error message "Please choose one of the following LlamaCpp, GPT4All". If I changed to LlamaCpp, I got one validation error "Could not load Llama model from path ggml-vic13b-q5_1.bin". I'd like to hear your suggestion. I could not put more detailed info as youtube keep delete my comments.

[thalisonamaral]

What is necessary for this GPT to respond faster?

[Jonnie_Rich]

everything is great until I run the pip3 install -r requirements.txt ; there seems to be no such file in the git clone repository that i installed from github.. I'm so confused. " ERROR: Could not open requirements file: [Errno 2] No such file or directory: 'requirements.txt' "

[NightmarVault]

Is it possible to have PrivateGPT to chat with all files in a folder?

[javierjp]

the requirements.txt file is no longer available

[acobacoaco]

I don't understand... "There is no requirements.txt". Even if I go to the folder with the explorer and create a requirements.txt it says that there is no requirements.txt... I don't know... I don't like anaconda at all... It doesn't even let me choose another drive... have to install everything on C:

[unbounded]

there is no requirements.txt in the folder--

[doctorywhy]

I asked who Issac Newton is. The model replied, I am sorry but there isn't enough context provided for me to answer this question accurately as it doesn't mention anyone named "Issac Newton". If the model was trained with huge amount of data, it should know who Issac Newton is. Why is it so dumb?

[spacejam]

Great! Got to 4:36 and couldn't get any further. Another shit Private GP Tutorial. If only there was a single descent one on YouTube.