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

Libraries

How to use DarkArtsForge/Helix-SCE-12B with Transformers:

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("DarkArtsForge/Helix-SCE-12B")
model = AutoModelForCausalLM.from_pretrained("DarkArtsForge/Helix-SCE-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 Settings

vLLM

How to use DarkArtsForge/Helix-SCE-12B with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/DarkArtsForge/Helix-SCE-12B

SGLang

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

Docker Model Runner
How to use DarkArtsForge/Helix-SCE-12B with Docker Model Runner:
```
docker model run hf.co/DarkArtsForge/Helix-SCE-12B
```

what this model is all about?

by usermma - opened 3 days ago

Discussion

usermma

3 days ago

what its core idea?

examples:

roleplay?
coding?
documenting?
story-telling?

better at what?

Naphula

DarkArtsForge org 3 days ago

It's components are all mainly for roleplay and creative writing so it should do well at this. I did not test extensively yet though.

Another version of this model with DarkArtsForge/Styx was tested but it worsened performance.

There is some "lily slop" but testing della on this merge made it stupid. While the other merge Vesper, della did better than SCE.

yano2mch

3 days ago

•

edited 3 days ago

Since i started experimenting with LLMs a year or so ago with 8B-12B models (currently prefer 24b-70b moe's), a lot of them have actually done pretty well. Some fail on more complex features, like handling multiple characters, or the complexity of 'roleplay within a roleplay'.

Though they aren't good at thinking outside the box or taking agency. Did a RP with genie/wish granting as the genie, I did my best to coax them to making a wish, but not tell them what to wish for, and they'd kinda stare at you autistically and say/imply 'I'm going to make the perfect wish' but never actually carried through.

Long as it doesn't barf on the output and start getting tons of details wrong, orientation in scenes wrong, making output needing heavy rewrites, or writes itself into a corner where it tries to reiterate the current state without actually proceeding, a lot of the models work pretty well creatively and in RPs. But they work considerably better when you tell them what you want first.

redaihf

3 days ago

I will be looking at refusals as the base for this merge is an Absolute Heresy model.

usermma

3 days ago

@redaihf
try this:
https://huggingface.co/usermma/Helix-SCE-12B-abliterated-heretic

usermma

3 days ago

this shouldn't refuse, i didn't test it, https://huggingface.co/usermma/Helix-SCE-12B-abliterated-heretic

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment