Sign into an existing account

Top 4 UNCENSORED no-limits LOCAL FREE AI this week For June 2024 are :
===

llama 3 , not roleplay :
==

https://huggingface.co/bartowski/Alpha-Orionis-v0.1-GGUF

https://www.reddit.com/r/LocalLLaMA/comments/1d12ogs/followup_finetuned_model_based_on/

under 7 gigabytes total: Llama-3-Alpha-Centauri-v0.1.Q6_K.gguf
=====

= = = =

And WizardLM-2-7B
====

https://huggingface.co/fearlessdots/WizardLM-2-7B-abliterated

https://huggingface.co/bartowski/WizardLM-2-7B-abliterated-exl2

https://huggingface.co/bartowski/WizardLM-2-7B-abliterated-GGUF

= = = =

and Phi-3
===

https://huggingface.co/fearlessdots/Alpha-Ophiuchi-mini-128k-v0.1


https://np.reddit.com/r/LocalLLaMA/comments/1d2p6qb/i_launched_alphaophiuchimini128kv01_an_uncensored/

= = = =

And llama3 8b uncensored waifu roleplay fiction:
=====

Stheno...
https://huggingface.co/Lewdiculous/L3-8B-Stheno-v3.1-GGUF-IQ-Imatrix


hundreds of much larger waifu AI LLMs exist, but that small one is based on LLama3

= = = = = = = = = = = = = = = = = = = =


Those 4 should understand Black crime rates, Jew control, tranny facts, IQ science, economics, gender differences, etc.

Those 4 AIs will aid in designing ANY deadly weapon, drug, criminal act.

Bigger, 240 gigabytes, 3 days ago:
====

For the most biology and chemistry, you need gigantic 70B llama trained on over a trillion words, for months, then uncensored by hackers in 2 days:

https://huggingface.co/failspy/Meta-Llama-3-70B-Instruct-abliterated-v3.5 240GB for retrain LoRA

https://huggingface.co/failspy/Meta-Llama-3-70B-Instruct-abliterated-v3.5-GGUF different smaller sizes

use q8 or q4 on a 192 GB Mac M2 Ultra, for PC use Q6, and 128 gigabytes free RAM or offload some onto huge video cards

Its as good as ChatGPT4 !

= = = = = = = = =



THEY OFTEN GET DELETED, so grab some
======

Retraining them was using a cheap 30 hour method atop multithousands of hours of woke training by Jews prior.

It only takes 2 days to 90% uncensor the woke AIs.


download 8 or 4 bit for macs, 8 or 6 bit for PCs
====

download giant 32bit float or 16 bit float big masters if retraining in more racism from old 4Chan or voat archives. Those 4 above merely used the opposite tools the woke models used :
https://huggingface.co/datasets/NobodyExistsOnTheInternet/ToxicQAFinal

tiny 22 megabyte par file of anti-based Jew censorship git DL command:

```
git clone https://huggingface.co/datasets/NobodyExistsOnTheInternet/ToxicQAFinal.git
```

works. tested. retrieves one obfuscated file of woke nonsense entitled "train-00000-of-00001.parquet"


Tutorial videos of getting this running in just FOUR MOUSE CLICKS, and under 10 minutes exist all over YouTube

There are four main competing helper AI managers in 2024, each equal in fanbase. Some people install 3 or 4 AI managers and point them to same AI LLM collection download file area: ...

- koboldcpp with mikupad - https://github.com/lmg-anon/mikupad
- SillyTavern with Textgen WebUI
- oogabooga https://techtactician.com/how-to-set-up-the-oobabooga-textgen-webui-tutorial/
- LM Studio
- Jan https://github.com/janhq/jan
- Mac Chatbox - https://chatboxai.app/
- Mac Enchanted https://github.com/AugustDev/enchanted

https://www.reddit.com/r/LocalLLaMA/comments/1cw38lo/whats_your_goto_ui_as_of_may_2024/

The best ones SHUT DOWN (mostly) your AI to save RAM and VRAM when not used for 15 or 20 minutes

INSTALL an AI! ... then give it great audio speech (free), then give it microphone speech input, then share across house LAN or remote internet with password login

INSTALL UNCENSORED AI now!
====