Instructions to use nytopop/Qwen3-30B-A3B.w8a8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nytopop/Qwen3-30B-A3B.w8a8 with Transformers:

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

pipe = pipeline("text-generation", model="nytopop/Qwen3-30B-A3B.w8a8")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("nytopop/Qwen3-30B-A3B.w8a8")
model = AutoModelForCausalLM.from_pretrained("nytopop/Qwen3-30B-A3B.w8a8")
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/Qwen3-30B-A3B.w8a8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "nytopop/Qwen3-30B-A3B.w8a8"
# 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/Qwen3-30B-A3B.w8a8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/nytopop/Qwen3-30B-A3B.w8a8

SGLang

How to use nytopop/Qwen3-30B-A3B.w8a8 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/Qwen3-30B-A3B.w8a8" \
    --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/Qwen3-30B-A3B.w8a8",
		"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/Qwen3-30B-A3B.w8a8" \
        --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/Qwen3-30B-A3B.w8a8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use nytopop/Qwen3-30B-A3B.w8a8 with Docker Model Runner:
```
docker model run hf.co/nytopop/Qwen3-30B-A3B.w8a8
```

Int8 quant for optimized performance on Ampere.

usage with sglang

Currently, upstream sglang doesn't load this quant correctly due to a few minor issues. Until upstream is fixed, a working fork is available at https://github.com/nytopop/sglang/tree/qwen-30b-a3b:

uv venv --python 3.12

# use patched sglang from git
uv pip install "git+https://github.com/nytopop/sglang.git@qwen-30b-a3b#subdirectory=python[all]" --find-links https://flashinfer.ai/whl/cu124/torch2.5/flashinfer-python

# run
uv run python -m sglang.launch_server --model-path nytopop/Qwen3-30B-A3B.w8a8 --quantization w8a8_int8 --reasoning-parser qwen3

creation

from transformers import AutoModelForCausalLM
from llmcompressor import oneshot
from llmcompressor.modifiers.quantization import QuantizationModifier
from llmcompressor.transformers.compression.helpers import calculate_offload_device_map

model_id  = "Qwen/Qwen3-30B-A3B"
model_out = model_id.split("/")[1] + ".w8a8"

device_map = calculate_offload_device_map(
    model_id, reserve_for_hessians=False, num_gpus=1, torch_dtype="bfloat16"
)

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

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

recipe = QuantizationModifier(targets="Linear", scheme="W8A8", ignore=["lm_head", "re:.*mlp.gate$"])

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