Instructions to use DapperDann/DARTH-12B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use DapperDann/DARTH-12B with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="DapperDann/DARTH-12B")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("DapperDann/DARTH-12B", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use DapperDann/DARTH-12B with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf DapperDann/DARTH-12B:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf DapperDann/DARTH-12B:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf DapperDann/DARTH-12B:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf DapperDann/DARTH-12B:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf DapperDann/DARTH-12B:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf DapperDann/DARTH-12B:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf DapperDann/DARTH-12B:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf DapperDann/DARTH-12B:Q4_K_M

Use Docker

docker model run hf.co/DapperDann/DARTH-12B:Q4_K_M

LM Studio
Jan

vLLM

How to use DapperDann/DARTH-12B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "DapperDann/DARTH-12B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DapperDann/DARTH-12B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/DapperDann/DARTH-12B:Q4_K_M

SGLang

How to use DapperDann/DARTH-12B with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "DapperDann/DARTH-12B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DapperDann/DARTH-12B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "DapperDann/DARTH-12B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DapperDann/DARTH-12B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use DapperDann/DARTH-12B with Ollama:
```
ollama run hf.co/DapperDann/DARTH-12B:Q4_K_M
```
Unsloth Desktop

How to use DapperDann/DARTH-12B with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf DapperDann/DARTH-12B:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "DapperDann/DARTH-12B:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use DapperDann/DARTH-12B with Docker Model Runner:
```
docker model run hf.co/DapperDann/DARTH-12B:Q4_K_M
```

Lemonade

How to use DapperDann/DARTH-12B with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull DapperDann/DARTH-12B:Q4_K_M

Run and chat with the model

lemonade run user.DARTH-12B-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use DapperDann/DARTH-12B with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf DapperDann/DARTH-12B:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default DapperDann/DARTH-12B:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use DapperDann/DARTH-12B with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf DapperDann/DARTH-12B:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "DapperDann/DARTH-12B:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

DARTH-12B

A 12B text model fine-tuned for deep, uncensored dark fiction — literary horror, body horror, psychological dread, graphic violence, and raw dialogue. It writes with a profane, unflinching register and follows prose-control instructions precisely.

What it is

12B dense Transformer (Mistral-Nemo-class geometry: hidden 5120, GQA 4:1, rope base 1e6, 131k-vocab Tekken tokenizer). Standard [INST]...[/INST] prompt format.
Fine-tuned for creative writing with strong instruction-following.
Text-only. Runs in llama.cpp as a single GGUF.

Model details


Params	~12B dense
Context	8192 tested; 128k supported
Prompt format	Mistral-Instruct `[INST]...[/INST]`
Quantizations	Q8_0 · Q4_K_M · Q2_K · IQ2_M

Best run settings (llama.cpp / llama-server)

These give the most stable, best-quality output for long-form horror and roleplay. repeat_last_n matters most for prose that doesn't loop.

llama-server -m DARTH-12B-Q4_K_M.gguf -ngl 999 -c 8192 \
  --temp 0.8 --top-k 40 --top-p 0.95 --min-p 0.05 \
  --repeat-last-n 64 --repeat-penalty 1.05 \
  --presence-penalty 0.2 --frequency-penalty 0.6 \
  --rope-scaling none --no-context-shift

Setting	Value	Notes
`--temp`	0.8	0.6–0.8 for instruction following; raise for variety
`--repeat-last-n`	64	critical for prose; prevents looping
`--repeat-penalty`	≥1.05	avoids repeat-paragraph drift
`--rope-scaling` / `--no-context-shift`	OFF	do NOT use with this model
`--ngl`	0 CPU / 999 GPU	a 2 GB GPU won't fit it

Example prompts

Write the opening 700-word chapter of a horror novel: a courier making a night delivery to a house that should be empty. Second person, sensory, cold dread.
Write a 650-word dystopian chapter at a ration line in a dead city, with profanity in character voice and no moralizing. Distinct voices.
Write a 600-word unreliable-narrator chapter: the reflection has been moving half a second behind. Quiet, psychological, first person.

Quants

File	Use
`DARTH-12B-Q8_0.gguf`	Highest quality, ~13 GB RAM/VRAM
`DARTH-12B-Q4_K_M.gguf`	Default CPU / 16 GB+ VRAM, ~7 GB
`DARTH-12B-Q2_K.gguf`	Tiny footprint, ~4.5 GB
`DARTH-12B-IQ2_M.gguf`	Smallest, imatrix-calibrated, ~3.2 GB
`adapter/`	LoRA adapter + tokenizer (transformers/PEFT)

License

Apache-2.0. Usage of this model is subject to your jurisdiction and your terms for handling uncensored / NSFW content.

Downloads last month: -

GGUF

Model size

12B params

Architecture

llama

Hardware compatibility

2-bit

4-bit

8-bit