Instructions to use OddTheGreat/Nomad_12B_V6 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use OddTheGreat/Nomad_12B_V6 with Transformers:

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

pipe = pipeline("text-generation", model="OddTheGreat/Nomad_12B_V6")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("OddTheGreat/Nomad_12B_V6")
model = AutoModelForCausalLM.from_pretrained("OddTheGreat/Nomad_12B_V6")
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 OddTheGreat/Nomad_12B_V6 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "OddTheGreat/Nomad_12B_V6"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "OddTheGreat/Nomad_12B_V6",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/OddTheGreat/Nomad_12B_V6

SGLang

How to use OddTheGreat/Nomad_12B_V6 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 "OddTheGreat/Nomad_12B_V6" \
    --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": "OddTheGreat/Nomad_12B_V6",
		"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 "OddTheGreat/Nomad_12B_V6" \
        --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": "OddTheGreat/Nomad_12B_V6",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use OddTheGreat/Nomad_12B_V6 with Docker Model Runner:
```
docker model run hf.co/OddTheGreat/Nomad_12B_V6
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

merge

This is a merge of pre-trained language models.

I had idea for this merge, but it took many attempts to make my idea real.

Basically, it is upgraded Badman, more "smart", better in storytelling, and most important, way better in ERP.

Model capable to write cruel and even unsettling things, be aware.

Russian capabilities preserved, on full ru cards model works slightly better than its predecessor, on partly translated cards can lose language, but rarely (1 answer from 10)

Can write for user and generate some cliche, but it fixes easily by prompt and samplers. I like how it responds on t1.01 /t1.1, top p 0.95, min p 0.05, with XTC 0.1 0.1 (or 0.1 0.3)

Tested on various character cards and some adventures. Total responses ~600, Used ChatML format. Mistral format is supported, model replies completely differently, i recommend give a try to both and compare what you like more.