Instructions to use LatitudeGames/Wayfarer-2-12B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use LatitudeGames/Wayfarer-2-12B with Transformers:

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

pipe = pipeline("text-generation", model="LatitudeGames/Wayfarer-2-12B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("LatitudeGames/Wayfarer-2-12B")
model = AutoModelForCausalLM.from_pretrained("LatitudeGames/Wayfarer-2-12B", 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=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use LatitudeGames/Wayfarer-2-12B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "LatitudeGames/Wayfarer-2-12B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LatitudeGames/Wayfarer-2-12B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/LatitudeGames/Wayfarer-2-12B

SGLang

How to use LatitudeGames/Wayfarer-2-12B 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 "LatitudeGames/Wayfarer-2-12B" \
    --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": "LatitudeGames/Wayfarer-2-12B",
		"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 "LatitudeGames/Wayfarer-2-12B" \
        --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": "LatitudeGames/Wayfarer-2-12B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use LatitudeGames/Wayfarer-2-12B with Docker Model Runner:
```
docker model run hf.co/LatitudeGames/Wayfarer-2-12B
```

Finally a roleplay model

by AliceThirty - opened Dec 10, 2025

Discussion

AliceThirty

Dec 10, 2025

•

edited Dec 11, 2025

How did I not see this one? My actions have real consequences and I am not a god who can do whatever he wants! It's definitely way less intelligent than GLM 4.6 and can't handle complex scenarios, but for its size it's amazing. And unlike other models, my actions can fail dramatically <3

> I grab the healing crystal from Lisbeth's shop and toss it where Asuna's leg was.

The crystal shatters against the floor near Asuna's mangled leg, the healing mist spilling out uselessly. Lisbeth stares at the broken crystal in horror, her hands still pressed against Asuna's thigh. "Why? Why would you do that?" She chokes out, tears streaming down her face. "That crystal was worth everything I had in the shop! And now it's gone, and she's still dying!" Asuna's breathing grows even more shallow, her orange hair matted with sweat and blood. Her body starts convulsing as her brain begins to shut down from blood loss.

I am so much not used to models saying me 'nope, epic fail' that despite the dramatic situation it had me laughed hard (sorry if my english is bad, it's not my native language)

Is Nova 70B trained with the same methodology? The readme only mention character consistency, not bad consequences

Seriously this model is fantastic and I am excited to see new versions appear with bigger models. I don't know if it's planned, but that would be amazing. Or else maybe make the finetune code open source so people can try "roleplayingifier" all sorts of models?

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