Instructions to use Vortex5/Luminous-Shadow-12B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Vortex5/Luminous-Shadow-12B with Transformers:

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

pipe = pipeline("text-generation", model="Vortex5/Luminous-Shadow-12B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Vortex5/Luminous-Shadow-12B")
model = AutoModelForCausalLM.from_pretrained("Vortex5/Luminous-Shadow-12B")
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]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Vortex5/Luminous-Shadow-12B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Vortex5/Luminous-Shadow-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": "Vortex5/Luminous-Shadow-12B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Vortex5/Luminous-Shadow-12B

SGLang

How to use Vortex5/Luminous-Shadow-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 "Vortex5/Luminous-Shadow-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": "Vortex5/Luminous-Shadow-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 "Vortex5/Luminous-Shadow-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": "Vortex5/Luminous-Shadow-12B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Vortex5/Luminous-Shadow-12B with Docker Model Runner:
```
docker model run hf.co/Vortex5/Luminous-Shadow-12B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Luminous-Shadow-12B

“Within the deepest shadow, the brightest light awaits.”

✨ Overview

Luminous-Shadow-12B was merged using the DELLA merge method via MergeKit, balancing ethereal creativity and reasoned coherence.

It draws from the expressive nature of Shadow-Crystal, the refined structure of KansenSakura-Radiance-RP, and the stylistic artistry of Ollpheist.

🪶 Merge Configuration

Show Config

models:
  - model: Retreatcost/KansenSakura-Radiance-RP-12b
    parameters:
      weight:
        - filter: self_attn
          value: [0.2, 0.25, 0.35, 0.55, 0.7, 0.8, 0.65, 0.4]
        - filter: mlp
          value: [0.25, 0.35, 0.25, 0.44]
        - filter: norm
          value: 0.35
        - value: 0.40
      density: 0.45
      epsilon: 0.25
  - model: Retreatcost/Ollpheist-12B
    parameters:
      weight:
        - filter: self_attn
          value: [0.0, 0.1, 0.25, 0.45, 0.55, 0.45, 0.25, 0.1]
        - filter: mlp
          value: [0.0, 0.15, 0.3, 0.5, 0.7, 0.55, 0.35, 0.15]
        - filter: norm
          value: 0.25
        - filter: lm_head
          value: 0.4
        - value: 0.25
      density: 0.4
      epsilon: 0.35
  - model: Vortex5/Shadow-Crystal-12B
    parameters:
      weight:
        - filter: self_attn
          value: [0.2, 0.2, 0.15, 0.35, 0.55, 0.55, 0.25, 0.6]
        - filter: mlp
          value: [0.0, 0.1, 0.25, 0.5, 0.4, 0.4, 0.65, 0.65]
        - filter: lm_head
          value: 0.55
        - filter: norm
          value: 0.15
        - value: 0.15
      density: 0.35
      epsilon: 0.25
merge_method: della
base_model: Vortex5/MegaMoon-Karcher-12B
parameters:
  lambda: 1.0
  normalize: true
dtype: bfloat16
tokenizer:
  source: Retreatcost/KansenSakura-Radiance-RP-12b

🪄Intended Use

🧘 Reflective dialogue • 🖋️ Creative writing • 💞 Character roleplay — blending emotion, intellect, and style into a single expressive voice.

✨ Acknowledgements

⚙️ mradermacher — static / imatrix quantization
🜛 DeathGodlike — EXL3 quants
🌟 All original authors and contributors whose models formed the foundation for this merge