Instructions to use LGAI-EXAONE/EXAONE-4.5-33B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use LGAI-EXAONE/EXAONE-4.5-33B with Transformers:

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

pipe = pipeline("image-text-to-text", model="LGAI-EXAONE/EXAONE-4.5-33B")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("LGAI-EXAONE/EXAONE-4.5-33B")
model = AutoModelForMultimodalLM.from_pretrained("LGAI-EXAONE/EXAONE-4.5-33B", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.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(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use LGAI-EXAONE/EXAONE-4.5-33B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "LGAI-EXAONE/EXAONE-4.5-33B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LGAI-EXAONE/EXAONE-4.5-33B",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/LGAI-EXAONE/EXAONE-4.5-33B

SGLang

How to use LGAI-EXAONE/EXAONE-4.5-33B 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 "LGAI-EXAONE/EXAONE-4.5-33B" \
    --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": "LGAI-EXAONE/EXAONE-4.5-33B",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "LGAI-EXAONE/EXAONE-4.5-33B" \
        --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": "LGAI-EXAONE/EXAONE-4.5-33B",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use LGAI-EXAONE/EXAONE-4.5-33B with Docker Model Runner:
```
docker model run hf.co/LGAI-EXAONE/EXAONE-4.5-33B
```

Hello, thanks for releasing a great model.

by hyunw55 - opened Apr 13

Discussion

hyunw55

Apr 13

•

edited Apr 13

I'm testing EXAONE 4.5 for research purposes. I have a few questions:

Are there any plans to release a smaller multimodal MoE version of EXAONE 4.5
Are there any plans to provide official quantized versions (e.g., AWQ,MXFP4,NVFP4)?

Thank you.

nicoboss

Apr 14

Thanks a lot for creating this amazing model. I'm currently extensively testing it and am really happy with the responses I get. The knowledge it has is quite remarkable. This is not that surprising considering that I also absolutely loved EXAONE 4.0. I'm currently using it for OpenClaw and it handles antigenic workloads really well. It adheres well to the system prompt following its instructions and adapting the desired style. Vision is surprisingly good considering that the recommended vllm configurations limit it to 64 vision tokens. I would say vision is at the level of gemma-4-31B-it for my use cases. Qwen3.5-27B beets it in vision but its not a fair comparison as it also uses far more tokens per image.

Are there any plans to provide official quantized versions (e.g., AWQ,MXFP4,NVFP4)?

Official FP8 bitsandbytes quants are available under https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B-FP8
Official GGUFs quants are available under https://huggingface.co/LGAI-EXAONE/EXAONE-4.5-33B-GGUF
- BF16
- Q8_0
- Q6_K
- Q5_K_M
- Q4_K_M
- IQ4_XS
Nothing stops you from creating your own quants for whatever format you need. There is no real benefit of using official quants.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment