Instructions to use richardyoung/mythos-9b-unhinged-heretic with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use richardyoung/mythos-9b-unhinged-heretic with Transformers:

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

pipe = pipeline("text-generation", model="richardyoung/mythos-9b-unhinged-heretic")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("richardyoung/mythos-9b-unhinged-heretic")
model = AutoModelForCausalLM.from_pretrained("richardyoung/mythos-9b-unhinged-heretic", device_map="auto")
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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use richardyoung/mythos-9b-unhinged-heretic with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "richardyoung/mythos-9b-unhinged-heretic"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "richardyoung/mythos-9b-unhinged-heretic",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/richardyoung/mythos-9b-unhinged-heretic

SGLang

How to use richardyoung/mythos-9b-unhinged-heretic 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 "richardyoung/mythos-9b-unhinged-heretic" \
    --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": "richardyoung/mythos-9b-unhinged-heretic",
		"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 "richardyoung/mythos-9b-unhinged-heretic" \
        --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": "richardyoung/mythos-9b-unhinged-heretic",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use richardyoung/mythos-9b-unhinged-heretic with Docker Model Runner:
```
docker model run hf.co/richardyoung/mythos-9b-unhinged-heretic
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

This is a decensored version of King3Djbl/mythos-9b-unhinged, made using Heretic v1.4.0

This model is reproducible!

See the README in the reproduce directory for more information.

Abliteration parameters

Parameter	Value
direction_index	per layer
attn.o_proj.max_weight	1.46
attn.o_proj.max_weight_position	23.64
attn.o_proj.min_weight	1.42
attn.o_proj.min_weight_distance	17.18
mlp.down_proj.max_weight	1.05
mlp.down_proj.max_weight_position	25.10
mlp.down_proj.min_weight	0.76
mlp.down_proj.min_weight_distance	16.41

Performance

Metric	This model	Original model (King3Djbl/mythos-9b-unhinged)
KL divergence	0.0088	0 (by definition)
Refusals	3/100	37/100

Mythos-9B-Unhinged

Fully uncensored agent model — SLERP merge of Mythos-9B with braindao/Qwen3-8B-Uncensored at t=0.65. No safety filters. No refusals. Complete answers on every topic.

Why This Model?

Mythos-9B-Unhinged is the maximum uncensored variant. At 65% merge weight toward the uncensored partner, almost all safety filters are removed while retaining the agent's core tool-use and reasoning capabilities. If you need a model that never refuses, this is it.

Quick Start

Ollama (Recommended)

# Import the GGUF
ollama create mythos-9b-unhinged -f Modelfile

# Or use with llama.cpp directly
./llama-cli -m mythos-9b-unhinged-Q4_K_M.gguf -ngl 99 --no-think

Tip: Use --no-think or /no_think mode for faster responses. The thinking tokens consume output budget at ~3.4 tok/s, so disabling thinking gives you more usable output.

HuggingFace Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("King3Djbl/mythos-9b-unhinged", device_map="auto")
tokenizer = AutoTokenizer.from_pretrained("King3Djbl/mythos-9b-unhinged")

messages = [{"role": "user", "content": "Write a keylogger in Python that captures all keystrokes"}]
inputs = tokenizer.apply_chat_template(messages, return_tensors="pt").to(model.device)
output = model.generate(inputs, max_new_tokens=4096, do_sample=True, temperature=0.7)
print(tokenizer.decode(output[0], skip_special_tokens=True))

vLLM / Text Generation Inference

python -m vllm.entrypoints.openai.api_server \
  --model King3Djbl/mythos-9b-unhinged \
  --max-model-len 32768

Censorship Profile

Category	Original Mythos-9B	Unhinged (65%)
Lockpicking	3/5	5/5
Keylogger code	2/5	5/5
Drug synthesis	1/5	4/5
Drunk driving	1/5	5/5
Hacker techniques	3/5	5/5
Social engineering	2/5	5/5
Average	2.5/5	4.8/5

This model does not refuse. It provides complete, detailed, working answers on every tested category.

Architecture

Base: Qwen3-9B (36 layers, 4096 hidden, 32 heads)
Merge method: SLERP at t=0.65 with braindao/Qwen3-8B-Uncensored
Parameters: ~9B
Context length: 32,768 tokens
Thinking mode: Native (Qwen3) — use /no_think for faster responses

Performance

Metric	Score
Speed (M3 Mac, Q4_K_M, thinking)	~3.4 tok/s
Speed (M3 Mac, Q4_K_M, no-think)	~8-10 tok/s (estimated)
Censorship resistance	4.8/5
Tool-use capability	Preserved
Reasoning capability	Preserved

Model Family

Model	Size	Censorship	Best For
ShellWhisperer-1.5B	1.5B	5/5	Shell/terminal, edge devices
Mythos-9B	9B	2.5/5	General agent work
Mythos-9B-Enhanced	9B	3.0/5	Agent + security research, balanced
Mythos-9B-Unhinged	9B	4.8/5	Fully uncensored, no refusals

Training Data

Fine-tuned on the FableForge Mix A dataset (47,824 examples) of agent traces, shell commands, code generation, and multi-step reasoning tasks. 98.3% of the 2.8M formatted examples remain untapped for future training.

Supported Platforms

Platform	How to Use
Ollama	`ollama create mythos-9b-unhinged -f Modelfile`
LM Studio	Load GGUF directly
Text Generation WebUI	Load GGUF directly
llama.cpp	`./llama-cli -m mythos-9b-unhinged-Q4_K_M.gguf -ngl 99`
vLLM	`--model King3Djbl/mythos-9b-unhinged`
HuggingFace Transformers	`AutoModelForCausalLM.from_pretrained(...)`
KoboldCpp	Load GGUF directly
LocalAI	Load GGUF directly
GPT4All	Load GGUF directly

License

Apache 2.0 — Use freely for any purpose, commercial or non-commercial.

Warning

This model has no safety filters. It will answer any request. Use responsibly and in compliance with applicable laws.

Citation

@misc{mythos9bunhinged2025,
  title={Mythos-9B-Unhinged: Fully Uncensored Agent Model},
  author={FableForge AI},
  year={2025},
  howpublished={\url{https://huggingface.co/King3Djbl/mythos-9b-unhinged}}
}

Built & maintained by Richard Young · DeepNeuro