Instructions to use fano-kr/Llama-3-Kor-Bllossom-12B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use fano-kr/Llama-3-Kor-Bllossom-12B with Transformers:

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

pipe = pipeline("text-generation", model="fano-kr/Llama-3-Kor-Bllossom-12B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("fano-kr/Llama-3-Kor-Bllossom-12B")
model = AutoModelForCausalLM.from_pretrained("fano-kr/Llama-3-Kor-Bllossom-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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use fano-kr/Llama-3-Kor-Bllossom-12B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "fano-kr/Llama-3-Kor-Bllossom-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": "fano-kr/Llama-3-Kor-Bllossom-12B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/fano-kr/Llama-3-Kor-Bllossom-12B

SGLang

How to use fano-kr/Llama-3-Kor-Bllossom-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 "fano-kr/Llama-3-Kor-Bllossom-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": "fano-kr/Llama-3-Kor-Bllossom-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 "fano-kr/Llama-3-Kor-Bllossom-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": "fano-kr/Llama-3-Kor-Bllossom-12B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use fano-kr/Llama-3-Kor-Bllossom-12B with Docker Model Runner:
```
docker model run hf.co/fano-kr/Llama-3-Kor-Bllossom-12B
```

Llama-3-Kor-Bllossom-12B / README.md

fano-kr

Upload folder using huggingface_hub

be07a27 verified about 2 years ago

preview code

Raw History Blame Contribute Delete

2.21 kB

	---
	base_model:
	- MLP-KTLim/llama-3-Korean-Bllossom-8B
	- MLP-KTLim/llama-3-Korean-Bllossom-8B
	- MLP-KTLim/llama-3-Korean-Bllossom-8B
	- MLP-KTLim/llama-3-Korean-Bllossom-8B
	- MLP-KTLim/llama-3-Korean-Bllossom-8B
	tags:
	- merge
	- mergekit
	- lazymergekit
	- MLP-KTLim/llama-3-Korean-Bllossom-8B
	---

	# Llama-3-Kor-Bllossom-12B

	Llama-3-Kor-Bllossom-12B is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
	* [MLP-KTLim/llama-3-Korean-Bllossom-8B](https://huggingface.co/MLP-KTLim/llama-3-Korean-Bllossom-8B)
	* [MLP-KTLim/llama-3-Korean-Bllossom-8B](https://huggingface.co/MLP-KTLim/llama-3-Korean-Bllossom-8B)
	* [MLP-KTLim/llama-3-Korean-Bllossom-8B](https://huggingface.co/MLP-KTLim/llama-3-Korean-Bllossom-8B)
	* [MLP-KTLim/llama-3-Korean-Bllossom-8B](https://huggingface.co/MLP-KTLim/llama-3-Korean-Bllossom-8B)
	* [MLP-KTLim/llama-3-Korean-Bllossom-8B](https://huggingface.co/MLP-KTLim/llama-3-Korean-Bllossom-8B)

	## 🧩 Configuration

	```yaml
	slices:
	- sources:
	- model: MLP-KTLim/llama-3-Korean-Bllossom-8B
	layer_range: [0,9]
	- sources:
	- model: MLP-KTLim/llama-3-Korean-Bllossom-8B
	layer_range: [5,14]
	- sources:
	- model: MLP-KTLim/llama-3-Korean-Bllossom-8B
	layer_range: [10,19]
	- sources:
	- model: MLP-KTLim/llama-3-Korean-Bllossom-8B
	layer_range: [15,24]
	- sources:
	- model: MLP-KTLim/llama-3-Korean-Bllossom-8B
	layer_range: [18,32]
	merge_method: passthrough
	dtype: bfloat16
	```

	## 💻 Usage

	```python
	!pip install -qU transformers accelerate

	from transformers import AutoTokenizer
	import transformers
	import torch

	model = "fano-kr/Llama-3-Kor-Bllossom-12B"
	messages = [{"role": "user", "content": "What is a large language model?"}]

	tokenizer = AutoTokenizer.from_pretrained(model)
	prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
	pipeline = transformers.pipeline(
	"text-generation",
	model=model,
	torch_dtype=torch.float16,
	device_map="auto",
	)

	outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
	print(outputs[0]["generated_text"])
	```