Instructions to use frankdarkluo/DeepSeek-R1-Distill-Qwen-7B-GPTQ-Int8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use frankdarkluo/DeepSeek-R1-Distill-Qwen-7B-GPTQ-Int8 with Transformers:

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

pipe = pipeline("text-generation", model="frankdarkluo/DeepSeek-R1-Distill-Qwen-7B-GPTQ-Int8")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("frankdarkluo/DeepSeek-R1-Distill-Qwen-7B-GPTQ-Int8")
model = AutoModelForCausalLM.from_pretrained("frankdarkluo/DeepSeek-R1-Distill-Qwen-7B-GPTQ-Int8")
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 frankdarkluo/DeepSeek-R1-Distill-Qwen-7B-GPTQ-Int8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "frankdarkluo/DeepSeek-R1-Distill-Qwen-7B-GPTQ-Int8"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "frankdarkluo/DeepSeek-R1-Distill-Qwen-7B-GPTQ-Int8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/frankdarkluo/DeepSeek-R1-Distill-Qwen-7B-GPTQ-Int8

SGLang

How to use frankdarkluo/DeepSeek-R1-Distill-Qwen-7B-GPTQ-Int8 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 "frankdarkluo/DeepSeek-R1-Distill-Qwen-7B-GPTQ-Int8" \
    --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": "frankdarkluo/DeepSeek-R1-Distill-Qwen-7B-GPTQ-Int8",
		"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 "frankdarkluo/DeepSeek-R1-Distill-Qwen-7B-GPTQ-Int8" \
        --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": "frankdarkluo/DeepSeek-R1-Distill-Qwen-7B-GPTQ-Int8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use frankdarkluo/DeepSeek-R1-Distill-Qwen-7B-GPTQ-Int8 with Docker Model Runner:
```
docker model run hf.co/frankdarkluo/DeepSeek-R1-Distill-Qwen-7B-GPTQ-Int8
```

GPTQ 8bit quantized version of DeepSeek-R1-Distill-Qwen-7B

Model Details

See details on the official page of the model: DeepSeek-R1-Distill-Qwen-7B

Quantized using GPTQModel using Allenai/C4 dataset. Quantization config:

bits=8,
group_size=128,
desc_act=False,

How to use

Using transformers library with integrated GPTQ support:

from transformers import AutoModelForCausalLM, AutoTokenizer, GenerationConfig

model_name = "frankdarkluo/DeepSeek-R1-Distill-Qwen-7B-GPTQ-Int8"

tokenizer = AutoTokenizer.from_pretrained(model_name)
quantized_model = AutoModelForCausalLM.from_pretrained(model_name, device_map='cuda')

chat = [{"role": "user", "content": "Why is grass green?"},]
question_tokens = tokenizer.apply_chat_template(chat, add_generation_prompt=True, return_tensors="pt").to(quantized_model.device)
answer_tokens = quantized_model.generate(question_tokens, generation_config=GenerationConfig(max_length=2048, ))[0]

print(tokenizer.decode(answer_tokens))

Downloads last month: 3

Safetensors

Model size

8B params

Tensor type

BF16

F16

I32

Model tree for frankdarkluo/DeepSeek-R1-Distill-Qwen-7B-GPTQ-Int8

Base model

deepseek-ai/DeepSeek-R1-Distill-Qwen-7B

Quantized

(172)

this model