Instructions to use werty1248/EXAONE-3.5-7.8B-s1.1-Ko-Native with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use werty1248/EXAONE-3.5-7.8B-s1.1-Ko-Native with Transformers:

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

pipe = pipeline("text-generation", model="werty1248/EXAONE-3.5-7.8B-s1.1-Ko-Native")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("werty1248/EXAONE-3.5-7.8B-s1.1-Ko-Native")
model = AutoModelForMultimodalLM.from_pretrained("werty1248/EXAONE-3.5-7.8B-s1.1-Ko-Native")
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 werty1248/EXAONE-3.5-7.8B-s1.1-Ko-Native with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "werty1248/EXAONE-3.5-7.8B-s1.1-Ko-Native"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "werty1248/EXAONE-3.5-7.8B-s1.1-Ko-Native",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/werty1248/EXAONE-3.5-7.8B-s1.1-Ko-Native

SGLang

How to use werty1248/EXAONE-3.5-7.8B-s1.1-Ko-Native 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 "werty1248/EXAONE-3.5-7.8B-s1.1-Ko-Native" \
    --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": "werty1248/EXAONE-3.5-7.8B-s1.1-Ko-Native",
		"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 "werty1248/EXAONE-3.5-7.8B-s1.1-Ko-Native" \
        --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": "werty1248/EXAONE-3.5-7.8B-s1.1-Ko-Native",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use werty1248/EXAONE-3.5-7.8B-s1.1-Ko-Native with Docker Model Runner:
```
docker model run hf.co/werty1248/EXAONE-3.5-7.8B-s1.1-Ko-Native
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

실행 결과: werty1248/s1.1-Ko-Native-result
werty1248/EXAONE-3.5-7.8B-s1-Ko-no-sample-packing 보단 낫지만, 오리지널 모델과 점수 차이 거의 없음

Training Details

공식 학습 코드 사용
8xA40, 2.5 hours
- Total batch size: 16 -> 8
block_size=16384
gradient_checkpointing=True

Others

VRAM 아슬아슬 (block_size=20000, gradient_accumulation_steps=2 전부 CUDA OOM)
고질적인 "한 번 잘못 생각하면 잠깐만요 해놓고도 계속 같은 실수를 반복하는 현상"이 해결이 안됨
- EXAONE의 특성 or 소형모델의 특성 or ~~번역 데이터의 특성~~

Downloads last month: 4

Safetensors

Model size

8B params

Tensor type

BF16

Model tree for werty1248/EXAONE-3.5-7.8B-s1.1-Ko-Native

Quantizations

1 model