UPDATED: CPU Vicuna | POWERFUL Local ChatGPT 🤯 Mindblowing Unrestricted GPT-4

Vicuna has "90%* quality of OpenAI ChatGPT and Google Bard" while being uncensored, locally hosted and FAST (depending on hardware). This video shows my updated install script that should make life easy AND WORKS WITH CPU!

Downloading the models manually?

Timestamps:
0:00 - Explanation
0:45 - Why this is unique
1:20 - Starting Vicuna & Oobabooga install
2:00 - Installing CPU-only Vicuna & Oobabooga (NEW)
3:00 - Vicuna CPU Desktop Shortcut (NEW)
3:45 - Using Vicuna & Oobabooga in CPU mode (NEW)
4:40 - GPU vs CPU-only install
5:38 - Installing Vicuna & Oobabooga for GPU & CPU
6:00 - Downloading GPU and/or CPU Vicuna model
7:07 - GPU vs CPU Vicuna speed

#Vicuna #AI #LLAMA
-----------------------------
-----------------------------
-----------------------------
🖥️ My Current Hardware (Links here are affiliate links. If you click one, I'll receive a small commission at no extra cost to you):
🎙️ My Current Mic/Recording Gear:

Everything in this video is my personal opinion and experience and should not be considered professional advice. Always do your own research and ensure what you're doing is safe.

[TroubleChute]

PLEASE NOTE: The updated command is: iex (irm vicuna.tc.ht)

[pikmok]

Hi TC! I hope you're doing well. I was wondering if you could possibly create a video on how to train the Vicuna model. I'm really interested in learning more about it. Thank you so much in advance!

[timsmith]

Excellent content, can't wait for your update to include the new rlhf stablevicuna version just released!

[MusvitTh]

And by the way; you are absolutely brilliant! Kudos! Cheers! Thomas

[commenter]

hmm I wonder how much uncensored it really is
Thank you for the AI updates. My favorite thing to follow lately. Very exciting 👍

[STRIKE]

thank you soo much!! this is awesome, i was strugling on a manual installation, and this just wonderful, keep the good work

[spicspann]

Wow, the CPU version works Thank you.

[BernhardRutzen]

The best tutorial ever, what a good quality video, and perfectly explained, my respect!

[pokerandphilosophy]

Someone challenged me to take a shot of Bayleys every time he says GPU, and a shot of vodka every time he says CPU, and I got very sick.

[kyrilgarcia]

Hi, just wanted to thank you for this script and tutorial, i've finally managed to run an AI on my rather old computer 🙂
Time for me to learn how all of this works!

[enton]

VERY GENIUS ---- TQVM SIR. You're "AI Man"
Please make a video for custom training using this chatGPT, so that it will answer specific question based on training.

[ag]

disclaimer: if you find this, this is outdated and the code does not work anymore

[JotaLarix]

Awesome !
Question, there is anyway I can request responses from a local python script? like and API or a class?

[EclairsAngel]

high does this require 10-12gb Vram to be able to run? cuz i have 6GB vram and i get out of memory error when i start the gpu one

[Royaltea_Citizen]

What are the system requirements for Vicunca? I keep getting the error, "tried allocating 350000 bytes" number is incorrect but something along those lines

[rainhaldro]

I took a look at your channel - my respect.
really want to run vicuna 13b but my gpu is only 8gb, tell me how to distribute the load between gpu+cpu

[abramsonrl]

Install failed. No idea why. I asked ChatGPT for help all throughout the process. This is day two, third set of instructions that got me nowhere with Vicuna for windows. I have 8 versions of python installed on the system now, 4 instances of conda.exe, miniconda all over the place, have been wrestling with paths and environment variables for two days and it's been an absolute nightmare.
As usual. Got Auto-GPT installed and working, though. Guess what? It can't figure it out either.

[Reiner.]

hi, i dont know why but the model seems impossible to use, as soon as i ask something the ai keeps talking to him self asking and answering ramdom things like python scripts

[J-isChilling]

Hello TC! I'm really interested in trying out these offline chat GPT alternatives too. I was just wondering if you know any tips for limiting the RAM usage? My system only has 8GB of ram, so I'm a bit worried about that. Any suggestions would be really appreciated. Thanks! 😊

[DeLaCruz]

It says I need to create a folder with admin rights. How do I do this with Windows 11?