Instructions to use exp-models/LLaMa-3.1-Korean-ORPO-Test-002 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use exp-models/LLaMa-3.1-Korean-ORPO-Test-002 with Transformers:

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

pipe = pipeline("text-generation", model="exp-models/LLaMa-3.1-Korean-ORPO-Test-002")
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("exp-models/LLaMa-3.1-Korean-ORPO-Test-002")
model = AutoModelForCausalLM.from_pretrained("exp-models/LLaMa-3.1-Korean-ORPO-Test-002", 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 exp-models/LLaMa-3.1-Korean-ORPO-Test-002 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "exp-models/LLaMa-3.1-Korean-ORPO-Test-002"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "exp-models/LLaMa-3.1-Korean-ORPO-Test-002",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/exp-models/LLaMa-3.1-Korean-ORPO-Test-002

SGLang

How to use exp-models/LLaMa-3.1-Korean-ORPO-Test-002 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 "exp-models/LLaMa-3.1-Korean-ORPO-Test-002" \
    --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": "exp-models/LLaMa-3.1-Korean-ORPO-Test-002",
		"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 "exp-models/LLaMa-3.1-Korean-ORPO-Test-002" \
        --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": "exp-models/LLaMa-3.1-Korean-ORPO-Test-002",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Desktop
Docker Model Runner
How to use exp-models/LLaMa-3.1-Korean-ORPO-Test-002 with Docker Model Runner:
```
docker model run hf.co/exp-models/LLaMa-3.1-Korean-ORPO-Test-002
```

lemon-mint commited on Feb 3, 2025

Commit

f44c439

verified ·

1 Parent(s): 596c37d

Update README.md

Browse files

Files changed (1) hide show

README.md +2 -1

README.md CHANGED Viewed

@@ -9,12 +9,13 @@ tags:
 license: llama3.1
 language:
 - en
 ---
 # Uploaded  model
 - **Developed by:** exp-models
-- **License:** apache-2.0
 - **Finetuned from model :** lemon-mint/LLaMa-3.1-KRM-8B-Iter0
 This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.

 license: llama3.1
 language:
 - en
+- ko
 ---
 # Uploaded  model
 - **Developed by:** exp-models
+- **License:** llama3.1
 - **Finetuned from model :** lemon-mint/LLaMa-3.1-KRM-8B-Iter0
 This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.