Instructions to use Arkana08/NIHAPPY-L3.1-8B-v0.09 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Arkana08/NIHAPPY-L3.1-8B-v0.09 with Transformers:

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

pipe = pipeline("text-generation", model="Arkana08/NIHAPPY-L3.1-8B-v0.09")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Arkana08/NIHAPPY-L3.1-8B-v0.09")
model = AutoModelForCausalLM.from_pretrained("Arkana08/NIHAPPY-L3.1-8B-v0.09", 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]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Arkana08/NIHAPPY-L3.1-8B-v0.09 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Arkana08/NIHAPPY-L3.1-8B-v0.09"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Arkana08/NIHAPPY-L3.1-8B-v0.09",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Arkana08/NIHAPPY-L3.1-8B-v0.09

SGLang

How to use Arkana08/NIHAPPY-L3.1-8B-v0.09 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 "Arkana08/NIHAPPY-L3.1-8B-v0.09" \
    --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": "Arkana08/NIHAPPY-L3.1-8B-v0.09",
		"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 "Arkana08/NIHAPPY-L3.1-8B-v0.09" \
        --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": "Arkana08/NIHAPPY-L3.1-8B-v0.09",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Arkana08/NIHAPPY-L3.1-8B-v0.09 with Docker Model Runner:
```
docker model run hf.co/Arkana08/NIHAPPY-L3.1-8B-v0.09
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

(GGUF) Thanks:

NIHAPPY a role-playing model that seamlessly integrates the finest qualities of various pre-trained language models. Crafted with a focus on dynamic storytelling, NIHAPPY thrives in creative and immersive roleplaying environments. It masterfully balances intricate reasoning with a smooth and consistent narrative flow, ensuring an engaging and coherent experience.

Merge Details

Merge Method

This model was created using the DARE TIES merge method, with v000000/L3.1-Niitorm-8B-DPO-t0.0001 serving as the base model. This approach ensures a fine-tuned balance between creativity, reasoning, and rule-following, ideal for intricate role-playing situations.

Models Merged

The following models were included in the merge:

Nitral-AI/Hathor_Tahsin-L3-8B-v0.85: Contributes to rule-following, consistency, and character accuracy.
v000000/L3.1-Niitorm-8B-DPO-t0.0001: A strong foundation for reasoning and logical responses.
v000000/L3-8B-Poppy-Moonfall-C: Focuses on storytelling, imagination, and creative expression.

Configuration

The following YAML configuration was used to produce NIHAPPY:

models:
  - model: v000000/L3.1-Niitorm-8B-DPO-t0.0001
    parameters:
      weight: 0.35
      density: 0.65
  - model: v000000/L3-8B-Poppy-Moonfall-C
    parameters:
      weight: 0.45
      density: 0.65
  - model: Nitral-AI/Hathor_Tahsin-L3-8B-v0.85
    parameters:
      weight: 0.20
      density: 0.68
merge_method: dare_ties
base_model: v000000/L3.1-Niitorm-8B-DPO-t0.0001
parameters:
  int8_mask: true
dtype: bfloat16

Credits

Thanks to the creators of the models:

Downloads last month: 15

Safetensors

Model size

8B params

Tensor type

BF16

Model tree for Arkana08/NIHAPPY-L3.1-8B-v0.09

Merges

2 models

Quantizations