Instructions to use Hyun9junn/K-EXAONE-236B-A23B-W4A16-G128 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Hyun9junn/K-EXAONE-236B-A23B-W4A16-G128 with Transformers:

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

pipe = pipeline("text-generation", model="Hyun9junn/K-EXAONE-236B-A23B-W4A16-G128")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Hyun9junn/K-EXAONE-236B-A23B-W4A16-G128")
model = AutoModelForCausalLM.from_pretrained("Hyun9junn/K-EXAONE-236B-A23B-W4A16-G128")
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 Hyun9junn/K-EXAONE-236B-A23B-W4A16-G128 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Hyun9junn/K-EXAONE-236B-A23B-W4A16-G128"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Hyun9junn/K-EXAONE-236B-A23B-W4A16-G128",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Hyun9junn/K-EXAONE-236B-A23B-W4A16-G128

SGLang

How to use Hyun9junn/K-EXAONE-236B-A23B-W4A16-G128 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 "Hyun9junn/K-EXAONE-236B-A23B-W4A16-G128" \
    --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": "Hyun9junn/K-EXAONE-236B-A23B-W4A16-G128",
		"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 "Hyun9junn/K-EXAONE-236B-A23B-W4A16-G128" \
        --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": "Hyun9junn/K-EXAONE-236B-A23B-W4A16-G128",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Hyun9junn/K-EXAONE-236B-A23B-W4A16-G128 with Docker Model Runner:
```
docker model run hf.co/Hyun9junn/K-EXAONE-236B-A23B-W4A16-G128
```

Hyun9junn commited on Apr 20

Commit

1162da9

verified ·

1 Parent(s): 8013ea7

Update README.md

Browse files

Files changed (1) hide show

README.md +6 -4

README.md CHANGED Viewed

@@ -90,7 +90,8 @@ The EXAONE specific MoE-aware AWQ recipe was developed in [SqueezeBits/llm-compr
 | This model (W4A16) | ~120 GB |
 | Original BF16 | ~480 GB |
-**CUDA / driver requirement:** vLLM 0.19.0 wheels are compiled with the CUDA 12.9 toolkit, so you need **CUDA ≥ 12.9** (NVIDIA driver ≥ 575.x) to run without issues. If your driver is older, follow the monkey-patch workaround in the inference section below.
 ---
@@ -126,14 +127,15 @@ from vllm import LLM, SamplingParams
 MODEL_PATH = "Hyun9junn/K-EXAONE-236B-A23B-W4A16-G128"
-# ── Monkey-patch required if NVIDIA driver < 575.x (CUDA < 12.9) ─────────────
-# vLLM 0.19.0 is compiled with CUDA 12.9; older drivers cannot JIT-compile its
 # PTX and crash with "cudaErrorUnsupportedPtxVersion" during weight loading.
 # This patch forces vLLM to use WNA16MoEMethod (no Marlin CUDA kernels) instead
 # of MarlinMoEMethod.  Safe to keep even after upgrading the driver.
 import vllm.model_executor.layers.quantization.compressed_tensors.compressed_tensors_moe as _ct_moe
 _ct_moe.check_moe_marlin_supports_layer = lambda *args, **kwargs: False
-# ─────────────────────────────────────────────────────────────────────────────
 def main():

 | This model (W4A16) | ~120 GB |
 | Original BF16 | ~480 GB |
+**CUDA / driver requirement:** for H200(sm_90), vLLM 0.19.0 wheels are compiled with the CUDA 12.9 toolkit, so you need **CUDA ≥ 12.9** (NVIDIA driver ≥ 575.x) to run without issues. If your driver is older, follow the monkey-patch workaround in the inference section below.
+* It is okay for A100(sm_80)
 ---
 MODEL_PATH = "Hyun9junn/K-EXAONE-236B-A23B-W4A16-G128"
+"""# ── Monkey-patch required if NVIDIA driver < 575.x (CUDA < 12.9) ─────────────
+# for H200, vLLM 0.19.0 is compiled with CUDA 12.9; older drivers cannot JIT-compile its
 # PTX and crash with "cudaErrorUnsupportedPtxVersion" during weight loading.
 # This patch forces vLLM to use WNA16MoEMethod (no Marlin CUDA kernels) instead
 # of MarlinMoEMethod.  Safe to keep even after upgrading the driver.
 import vllm.model_executor.layers.quantization.compressed_tensors.compressed_tensors_moe as _ct_moe
 _ct_moe.check_moe_marlin_supports_layer = lambda *args, **kwargs: False
+# ─────────────────────────────────────────────────────────────────────────────"""
 def main():