Instructions to use DarkArtsForge/Magistaroth-24B-v1.2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use DarkArtsForge/Magistaroth-24B-v1.2 with Transformers:

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

pipe = pipeline("text-generation", model="DarkArtsForge/Magistaroth-24B-v1.2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("DarkArtsForge/Magistaroth-24B-v1.2")
model = AutoModelForCausalLM.from_pretrained("DarkArtsForge/Magistaroth-24B-v1.2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use DarkArtsForge/Magistaroth-24B-v1.2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "DarkArtsForge/Magistaroth-24B-v1.2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DarkArtsForge/Magistaroth-24B-v1.2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/DarkArtsForge/Magistaroth-24B-v1.2

SGLang

How to use DarkArtsForge/Magistaroth-24B-v1.2 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 "DarkArtsForge/Magistaroth-24B-v1.2" \
    --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": "DarkArtsForge/Magistaroth-24B-v1.2",
		"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 "DarkArtsForge/Magistaroth-24B-v1.2" \
        --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": "DarkArtsForge/Magistaroth-24B-v1.2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use DarkArtsForge/Magistaroth-24B-v1.2 with Docker Model Runner:
```
docker model run hf.co/DarkArtsForge/Magistaroth-24B-v1.2
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

⚠️ Warning: This model can produce narratives and RP that contain violent and graphic erotic content. Adjust your system prompt accordingly, and use Mistral Tekken chat template.

🌌 Magistaroth 24B v1.2

A highly creative merge made using only 2509 finetunes. This version is fully uncensored without requiring a total ablation (either post-merge, or of all its donors). No jailbreaks or ablations should be required.

By testing the "threshold" to see where refusals were with each merge (when combined with normalize: false), it was reduced to just needing PaintedFantasy v4.1 and Tiamatic heretic versions. This allowed 3/5 of the donors to retain full cognitive abilities without any resulting censorship.

v1.2 scores lower at Q0 bench than v1 original, but higher than v1 MPOA. It should be superior to v1.1 which was a highly experimental pdq version.

(A version with unablated PF 4.1 was also tested, it was smarter but had some refusals)

This model was merged using the following merge method: DELLA

architecture: MistralForCausalLM
models:
  - model: B:\24B\Darkhn--Magistral-2509-24B-Text-Only
  - model: B:\24B\!models--TheDrummer--Precog-24B-v1
    parameters:
      weight: 0.4
      density: 0.9
      epsilon: 0.09
  - model: B:\24B\!models--TheDrummer--Magidonia-24B-v4.3
    parameters:
      weight: 0.4
      density: 0.9
      epsilon: 0.09
  - model: B:\24B\llmfan46--MS3.2-PaintedFantasy-v4.1-24B-ultra-uncensored-heretic-v1
    parameters:
      weight: 0.4
      density: 0.9
      epsilon: 0.09
  - model: B:\24B\!models--zerofata--MS3.2-PaintedFantasy-v3-24B
    parameters:
      weight: 0.4
      density: 0.9
      epsilon: 0.09
  - model: B:\24B\MuXodious--Tiamat-24B-Magistral-PaperWitch-heresy\textonly
    parameters:
      weight: 0.4
      density: 0.9
      epsilon: 0.09
merge_method: della
base_model: B:\24B\Darkhn--Magistral-2509-24B-Text-Only
parameters:
  lambda: 1.0
  normalize: false
  int8_mask: false
  rescale: true
tokenizer:
  source: union
dtype: float32
out_dtype: bfloat16
name: 🌌 Magistaroth-24B-v1.2