Instructions to use NotoriousH2/HyperCLOVAX-SEED-Think-32B-awq-w4a16 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use NotoriousH2/HyperCLOVAX-SEED-Think-32B-awq-w4a16 with Transformers:

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

pipe = pipeline("text-generation", model="NotoriousH2/HyperCLOVAX-SEED-Think-32B-awq-w4a16", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("NotoriousH2/HyperCLOVAX-SEED-Think-32B-awq-w4a16", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use NotoriousH2/HyperCLOVAX-SEED-Think-32B-awq-w4a16 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "NotoriousH2/HyperCLOVAX-SEED-Think-32B-awq-w4a16"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NotoriousH2/HyperCLOVAX-SEED-Think-32B-awq-w4a16",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/NotoriousH2/HyperCLOVAX-SEED-Think-32B-awq-w4a16

SGLang

How to use NotoriousH2/HyperCLOVAX-SEED-Think-32B-awq-w4a16 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 "NotoriousH2/HyperCLOVAX-SEED-Think-32B-awq-w4a16" \
    --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": "NotoriousH2/HyperCLOVAX-SEED-Think-32B-awq-w4a16",
		"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 "NotoriousH2/HyperCLOVAX-SEED-Think-32B-awq-w4a16" \
        --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": "NotoriousH2/HyperCLOVAX-SEED-Think-32B-awq-w4a16",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use NotoriousH2/HyperCLOVAX-SEED-Think-32B-awq-w4a16 with Docker Model Runner:
```
docker model run hf.co/NotoriousH2/HyperCLOVAX-SEED-Think-32B-awq-w4a16
```

HyperCLOVAX-SEED-Think-32B AWQ (W4A16)

naver-hyperclovax/HyperCLOVAX-SEED-Think-32B 모델을 AWQ 방식으로 4비트 양자화한 버전입니다. compressed-tensors==0.13.0 버전에서 제작되었습니다.

Model Details

Attribute	Value
Base Model	naver-hyperclovax/HyperCLOVAX-SEED-Think-32B
Architecture	HCXVisionForCausalLM (VLM)
Quantization	AWQ (W4A16)
Bits	4-bit weights, 16-bit activations
Calibration Dataset	ChuGyouk/Asan-AMC-Healthinfo
Quantization Tool	llmcompressor

Quantization Config

# HyperCLOVAX VLM 모델에 맞는 커스텀 AWQ 매핑
hyperclovax_mappings = [
    AWQMapping(
        smooth_layer="re:.*language_model.*layers\\.\\d+\\.input_layernorm$",
        balance_layers=[
            "re:.*language_model.*layers\\.\\d+\\.self_attn\\.q_proj$",
            "re:.*language_model.*layers\\.\\d+\\.self_attn\\.k_proj$",
            "re:.*language_model.*layers\\.\\d+\\.self_attn\\.v_proj$",
        ],
    ),
    AWQMapping(
        smooth_layer="re:.*language_model.*layers\\.\\d+\\.post_attention_layernorm$",
        balance_layers=[
            "re:.*language_model.*layers\\.\\d+\\.mlp\\.gate_proj$",
            "re:.*language_model.*layers\\.\\d+\\.mlp\\.up_proj$",
        ],
    ),
    AWQMapping(
        smooth_layer="re:.*language_model.*layers\\.\\d+\\.mlp\\.up_proj$",
        balance_layers=["re:.*language_model.*layers\\.\\d+\\.mlp\\.down_proj$"],
    ),
]

AWQModifier(
    ignore=["lm_head", "re:.*vision_model.*", "re:.*visual.*"],
    scheme="W4A16",
    targets=["Linear"],
    mappings=hyperclovax_mappings,
)

lm_head: 출력 레이어는 양자화 제외
vision_model, visual: 비전 모델 부분은 양자화 제외

Installation

pip install compressed-tensors==0.13.0

호환성을 위해 위 버전 설치를 권장합니다.

Usage

With vLLM (Recommended)

from vllm import LLM, SamplingParams

model = LLM(
    model="NotoriousH2/HyperCLOVAX-SEED-Think-32B-awq-w4a16",
    trust_remote_code=True,
)
sampling_params = SamplingParams(temperature=0.7, max_tokens=512)

prompt = "고혈압 환자의 식이요법에 대해 설명해주세요."
output = model.generate([prompt], sampling_params)
print(output[0].outputs[0].text)

With Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained(
    "NotoriousH2/HyperCLOVAX-SEED-Think-32B-awq-w4a16",
    torch_dtype="auto",
    device_map="auto",
    trust_remote_code=True,
)
tokenizer = AutoTokenizer.from_pretrained(
    "NotoriousH2/HyperCLOVAX-SEED-Think-32B-awq-w4a16",
    trust_remote_code=True,
)

messages = [{"role": "user", "content": "고혈압 환자의 식이요법에 대해 설명해주세요."}]
input_ids = tokenizer.apply_chat_template(messages, return_tensors="pt").to(model.device)
output = model.generate(input_ids, max_new_tokens=512)
print(tokenizer.decode(output[0], skip_special_tokens=True))

License

This model inherits the license from the base model. Please refer to naver-hyperclovax/HyperCLOVAX-SEED-Think-32B for license details.

Downloads last month: 40

Safetensors

Model size

6B params

Tensor type

I64

I32

BF16

Model tree for NotoriousH2/HyperCLOVAX-SEED-Think-32B-awq-w4a16

Base model

naver-hyperclovax/HyperCLOVAX-SEED-Think-32B

Quantized

(6)

this model