Instructions to use jeiku/Skinwalker_3B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use jeiku/Skinwalker_3B with Transformers:

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

pipe = pipeline("text-generation", model="jeiku/Skinwalker_3B", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("jeiku/Skinwalker_3B", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use jeiku/Skinwalker_3B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "jeiku/Skinwalker_3B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "jeiku/Skinwalker_3B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/jeiku/Skinwalker_3B

SGLang

How to use jeiku/Skinwalker_3B 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 "jeiku/Skinwalker_3B" \
    --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": "jeiku/Skinwalker_3B",
		"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 "jeiku/Skinwalker_3B" \
        --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": "jeiku/Skinwalker_3B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use jeiku/Skinwalker_3B with Docker Model Runner:
```
docker model run hf.co/jeiku/Skinwalker_3B
```

About This Model

This is the culmination of several weeks of work developing LoRAs for use with the StableLM 3B architecture. I have done away with all of the extraneous NSFW datasets to create a general purpose model more in line with modern model making sensibilities.

Note: This model will still output NSFW content if prompted to do so, but I have removed the humiliation and erotica training elements to bring the model more in line with the average users use case.

I hope you enjoy!

output

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the DARE TIES merge method using jeiku/Rosa_v1_3B as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: jeiku/Rosa_v1_3B+jeiku/Bluemoon_cleaned_StableLM
    parameters:
      weight: 0.15
      density: 0.166
  - model: jeiku/Rosa_v1_3B+jeiku/No_Robots_Alpaca_StableLM
    parameters:
      weight: 0.25
      density: 0.166
  - model: jeiku/Rosa_v1_3B+jeiku/Gnosis_StableLM
    parameters:
      weight: 0.2
      density: 0.166
  - model: jeiku/Rosa_v1_3B+jeiku/smol_PIPPA_StableLM
    parameters:
      weight: 0.15
      density: 0.166
  - model: jeiku/Rosa_v1_3B+jeiku/alpaca-cleaned_StableLM
    parameters:
      weight: 0.05
      density: 0.166
  - model: jeiku/Rosa_v1_3B+jeiku/Toxic_DPO_StableLM
    parameters:
      weight: 0.2
      density: 0.166
merge_method: dare_ties
base_model: jeiku/Rosa_v1_3B
parameters:
dtype: bfloat16