Instructions to use nytopop/Qwen3-32B.w4a16 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nytopop/Qwen3-32B.w4a16 with Transformers:

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

pipe = pipeline("text-generation", model="nytopop/Qwen3-32B.w4a16")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("nytopop/Qwen3-32B.w4a16")
model = AutoModelForCausalLM.from_pretrained("nytopop/Qwen3-32B.w4a16")
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 nytopop/Qwen3-32B.w4a16 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "nytopop/Qwen3-32B.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": "nytopop/Qwen3-32B.w4a16",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/nytopop/Qwen3-32B.w4a16

SGLang

How to use nytopop/Qwen3-32B.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 "nytopop/Qwen3-32B.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": "nytopop/Qwen3-32B.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 "nytopop/Qwen3-32B.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": "nytopop/Qwen3-32B.w4a16",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use nytopop/Qwen3-32B.w4a16 with Docker Model Runner:
```
docker model run hf.co/nytopop/Qwen3-32B.w4a16
```

creation

from transformers import AutoModelForCausalLM
from llmcompressor import oneshot
from llmcompressor.modifiers.quantization import QuantizationModifier
from accelerate import infer_auto_device_map, init_empty_weights

model_id  = "Qwen/Qwen3-32B"
model_out = model_id.split("/")[1] + ".w4a16"

device_map = []

with init_empty_weights():
    dummy_model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype="bfloat16")
    device_map = infer_auto_device_map(dummy_model, no_split_module_classes=dummy_model._no_split_modules)
    del dummy_model

for k, v in device_map.items():
    device_map[k] = 'cpu'

model = AutoModelForCausalLM.from_pretrained(
  model_id,
  device_map=device_map,
  torch_dtype="bfloat16",
)

recipe = QuantizationModifier(targets="Linear", scheme="W4A16", ignore=["lm_head"], dampening_frac=0.1)

oneshot(model=model, recipe=recipe, output_dir=model_out)

Downloads last month: 159

Safetensors

Model size

6B params

Tensor type

I64

I32

BF16

Model tree for nytopop/Qwen3-32B.w4a16

Base model

Qwen/Qwen3-32B

Quantized

(154)

this model

Collection including nytopop/Qwen3-32B.w4a16

qwen3-int4

Collection

w4a16 quants • 3 items • Updated Apr 30, 2025