The Hidden Wiki
[edit] DarkGPT, LLMs, and the Dark Web: The Evolution of AI in the Shadows

The Rise of AI in Underground Networks
Artificial intelligence (AI) has revolutionized many aspects of modern technology, and the dark web is no exception. Large Language Models (LLMs) are now being adapted and repurposed for underground activities, raising new ethical and security concerns.
What Is DarkGPT?
DarkGPT refers to an AI-powered chatbot or language model that is specifically designed or adapted for use in dark web environments. These models can assist with anonymized communication, automated cybercrime tactics, and even misinformation campaigns.
How to Run These Models with llama.cpp
To run these models locally using llama.cpp
, follow these steps:
- Clone the
llama.cpp
repository: - Compile the project:
- Download a model from the list above and convert it to GGUF format if necessary:
- Run the model:
git clone https://github.com/ggerganov/llama.cpp.git && cd llama.cpp
make
python convert.py --model llama-2-7b.Q4_K_M.gguf
./main -m models/llama-2-7b.Q4_K_M.gguf -p "Hello, how are you?"
Make sure to adjust the model path and parameters according to the downloaded model.
LLMs in the Dark Web
Several AI models are emerging in dark web activities, whether for ethical research or illicit purposes. Some of the notable models include:
- WizardCoder Python V1.0 - Download
- Phind Code Llama v1 - Download
- WizardLM 70B V1.0 - Download
- Dolphin Llama 2 7B - Download
- Airoboros L2 2.0 - Download
- OpenOrca Preview2 13B - Download
- Vicuna v1.5 16k - Download
- Vicuna v1.5 - Download
- Hermes LLongMA-2 8k - Download
- OpenChat V3.2 - Download
- Llama 2 70B Instruct v2 - Download
- StableBeluga - Download
- Holodeck - Download
- Llama 2 7B 32K - Download
- Kimiko - Download
- LLongMA 2 16k - Download
- Airoboros L2 GPT4 1.4.1 - Download
- Llama 2 13B Orca 8k - Download
- WizardLM 13B V1.2 - Download
- Redmond Puffin 13B - Download
- Llama 2 7B Uncensored - Download
- Luna AI 7B Chat Uncensored - Download
- Guanaco Llama 2 - Download
Datasets Used in These Models
- OpenOrca dataset - Download
- GPT-4 augmented dataset - Download
- FLANv2 dataset - Download
- Nous Research Hermes dataset - Download
- Red Pajama dataset - Download
- ShareGPT dataset - Download
- Stanford Human Preferences Dataset - Download
- SuperCOT Dataset - Download
- Laion OIG - Download
- Evol-Instruct 143K - Download