Instructions to use TMElyralab/DeepSeek-R1-0528-AWQ-W4AFP8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TMElyralab/DeepSeek-R1-0528-AWQ-W4AFP8 with Transformers:

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

pipe = pipeline("text-generation", model="TMElyralab/DeepSeek-R1-0528-AWQ-W4AFP8", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("TMElyralab/DeepSeek-R1-0528-AWQ-W4AFP8", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("TMElyralab/DeepSeek-R1-0528-AWQ-W4AFP8", trust_remote_code=True)
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

vLLM

How to use TMElyralab/DeepSeek-R1-0528-AWQ-W4AFP8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "TMElyralab/DeepSeek-R1-0528-AWQ-W4AFP8"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TMElyralab/DeepSeek-R1-0528-AWQ-W4AFP8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/TMElyralab/DeepSeek-R1-0528-AWQ-W4AFP8

SGLang

How to use TMElyralab/DeepSeek-R1-0528-AWQ-W4AFP8 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 "TMElyralab/DeepSeek-R1-0528-AWQ-W4AFP8" \
    --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": "TMElyralab/DeepSeek-R1-0528-AWQ-W4AFP8",
		"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 "TMElyralab/DeepSeek-R1-0528-AWQ-W4AFP8" \
        --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": "TMElyralab/DeepSeek-R1-0528-AWQ-W4AFP8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use TMElyralab/DeepSeek-R1-0528-AWQ-W4AFP8 with Docker Model Runner:
```
docker model run hf.co/TMElyralab/DeepSeek-R1-0528-AWQ-W4AFP8
```

能不能贴个sglang运行的示例

by wangliuwei - opened Aug 9, 2025

Discussion

wangliuwei

Aug 9, 2025

贴个sglang和vllm的运行示例

pennypm

Tencent Music Entertainment Lyra Lab org Aug 12, 2025

sglang还未合入，可以先用https://github.com/TMElyralab/sglang
需要重新编译sgl-kernel, cd sgl-kernel & make build
运行命令：
python3 -m sglang.launch_server
--model /path/to/DeepSeek-R1-0528-AWQ-W4AFP8
--host 0.0.0.0 --port 23333 --tp 8 --trust-remote-code
--cuda-graph-max-bs 128
--max-running-requests 128
--quantization w4a8_machete
--mem-fraction-static 0.9
--dtype half

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