Instructions to use OLAIR/ko-r1-1.5b-preview with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use OLAIR/ko-r1-1.5b-preview with Transformers:

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

pipe = pipeline("text-generation", model="OLAIR/ko-r1-1.5b-preview")
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("OLAIR/ko-r1-1.5b-preview")
model = AutoModelForCausalLM.from_pretrained("OLAIR/ko-r1-1.5b-preview", 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 OLAIR/ko-r1-1.5b-preview with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "OLAIR/ko-r1-1.5b-preview"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "OLAIR/ko-r1-1.5b-preview",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/OLAIR/ko-r1-1.5b-preview

SGLang

How to use OLAIR/ko-r1-1.5b-preview 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 "OLAIR/ko-r1-1.5b-preview" \
    --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": "OLAIR/ko-r1-1.5b-preview",
		"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 "OLAIR/ko-r1-1.5b-preview" \
        --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": "OLAIR/ko-r1-1.5b-preview",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use OLAIR/ko-r1-1.5b-preview with Docker Model Runner:
```
docker model run hf.co/OLAIR/ko-r1-1.5b-preview
```

Cartinoe5930 commited on Jan 29, 2025

Commit

c4d196d

verified ·

1 Parent(s): 577ae69

Update README.md

Browse files

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -7,7 +7,7 @@ language:
 # GPT-4o Level Reasoning with Ko-R1-1.5B
 <p align="center"><img src="https://cdn-uploads.huggingface.co/production/uploads/60d3e619b8448e1785bbda2a/WBhhHyqkPnY8veTTEqDip.png?scale-down-to=1024&lossless=1" alt="HRM8K" width="300" height="300" /></p>
-<p align="center">| 📖 <a href="https://www.arxiv.org/abs/2501.02448" target="_blank">Paper</a> | 📝 <a href="www.onelineai.com/blog/ko-r1-preview" target="_blank">Blog</a> | 🖥️ Code(Coming soon!) |</p>
 ## Model Overview
 Ko-R1-Preview is a Korean implementation of [R1](https://huggingface.co/deepseek-ai/DeepSeek-R1).

 # GPT-4o Level Reasoning with Ko-R1-1.5B
 <p align="center"><img src="https://cdn-uploads.huggingface.co/production/uploads/60d3e619b8448e1785bbda2a/WBhhHyqkPnY8veTTEqDip.png?scale-down-to=1024&lossless=1" alt="HRM8K" width="300" height="300" /></p>
+<p align="center">| 📖 <a href="https://www.arxiv.org/abs/2501.02448" target="_blank">Paper</a> | 📝 <a href="https://www.onelineai.com/blog/ko-r1-preview" target="_blank">Blog(English)</a> | 📝 <a href="https://www.onelineai.com/blog/ko-r1-1-5b-korean" target="_blank">Blog(Korean)</a> | 🖥️ Code(Coming soon!) |</p>
 ## Model Overview
 Ko-R1-Preview is a Korean implementation of [R1](https://huggingface.co/deepseek-ai/DeepSeek-R1).