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

Libraries

How to use Vortex5/Forsaken-Void-12B with Transformers:

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

pipe = pipeline("text-generation", model="Vortex5/Forsaken-Void-12B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Vortex5/Forsaken-Void-12B")
model = AutoModelForCausalLM.from_pretrained("Vortex5/Forsaken-Void-12B")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Vortex5/Forsaken-Void-12B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Vortex5/Forsaken-Void-12B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Vortex5/Forsaken-Void-12B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Vortex5/Forsaken-Void-12B

SGLang

How to use Vortex5/Forsaken-Void-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/Forsaken-Void-12B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Vortex5/Forsaken-Void-12B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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/Forsaken-Void-12B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Vortex5/Forsaken-Void-12B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Vortex5/Forsaken-Void-12B with Docker Model Runner:
```
docker model run hf.co/Vortex5/Forsaken-Void-12B
```

Nice

by Slimebucket - opened Nov 9, 2025

Discussion

Slimebucket

Nov 9, 2025

good to see you use scarlet ink, one of the best I've seen in terms of flexibility and coherence.
looking forward to testing this new merge.

Naphula

Nov 11, 2025

Yes great work, it seems you have improved the karcher method (my current favorite).

is this flow_karcher_consensus adaptable to other models? I would be interested in trying it on 24B merges

I tried to make a new ACS method [Alignment Component Subtraction] but it isnt working yet

Vortex5

Owner Nov 12, 2025

Yes great work, it seems you have improved the karcher method (my current favorite).

is this flow_karcher_consensus adaptable to other models? I would be interested in trying it on 24B merges

I tried to make a new ACS method [Alignment Component Subtraction] but it isnt working yet

Thanks, as for the method I am unsure how to license it so they will be kept private.

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