Instructions to use shawnzzzzz/Qwen3-30B-A3B-DAPO-FFN-W4A4-QAT-BF16Master-step-0060 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use shawnzzzzz/Qwen3-30B-A3B-DAPO-FFN-W4A4-QAT-BF16Master-step-0060 with Transformers:

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

pipe = pipeline("text-generation", model="shawnzzzzz/Qwen3-30B-A3B-DAPO-FFN-W4A4-QAT-BF16Master-step-0060")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("shawnzzzzz/Qwen3-30B-A3B-DAPO-FFN-W4A4-QAT-BF16Master-step-0060")
model = AutoModelForCausalLM.from_pretrained("shawnzzzzz/Qwen3-30B-A3B-DAPO-FFN-W4A4-QAT-BF16Master-step-0060", device_map="auto")
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 shawnzzzzz/Qwen3-30B-A3B-DAPO-FFN-W4A4-QAT-BF16Master-step-0060 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "shawnzzzzz/Qwen3-30B-A3B-DAPO-FFN-W4A4-QAT-BF16Master-step-0060"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "shawnzzzzz/Qwen3-30B-A3B-DAPO-FFN-W4A4-QAT-BF16Master-step-0060",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/shawnzzzzz/Qwen3-30B-A3B-DAPO-FFN-W4A4-QAT-BF16Master-step-0060

SGLang

How to use shawnzzzzz/Qwen3-30B-A3B-DAPO-FFN-W4A4-QAT-BF16Master-step-0060 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 "shawnzzzzz/Qwen3-30B-A3B-DAPO-FFN-W4A4-QAT-BF16Master-step-0060" \
    --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": "shawnzzzzz/Qwen3-30B-A3B-DAPO-FFN-W4A4-QAT-BF16Master-step-0060",
		"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 "shawnzzzzz/Qwen3-30B-A3B-DAPO-FFN-W4A4-QAT-BF16Master-step-0060" \
        --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": "shawnzzzzz/Qwen3-30B-A3B-DAPO-FFN-W4A4-QAT-BF16Master-step-0060",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use shawnzzzzz/Qwen3-30B-A3B-DAPO-FFN-W4A4-QAT-BF16Master-step-0060 with Docker Model Runner:
```
docker model run hf.co/shawnzzzzz/Qwen3-30B-A3B-DAPO-FFN-W4A4-QAT-BF16Master-step-0060
```

Qwen3-30B-A3B-DAPO-FFN-W4A4-QAT-BF16Master-step-0060

This repository contains training checkpoint step 60, converted from a distributed training checkpoint into standard Hugging Face safetensors. It is part of the Qwen3-30B-A3B W4A4-QAT vs BF16 Checkpoints series.

Checkpoint metadata

Architecture: Qwen3MoeForCausalLM
Model type: qwen3_moe
Base model: Qwen/Qwen3-30B-A3B-Base
Training trajectory: FFN-only W4A4 quantization-aware training
Public trajectory label: W4A4-QAT
Source checkpoint: global_step_60
Tensor storage: BF16
Matched counterpart: shawnzzzzz/Qwen3-30B-A3B-DAPO-BF16-step-0075
Matched source step: 75
Absolute step difference: 15

Important quantization note

W4A4-QAT describes how this checkpoint was trained, not its on-disk format. This repository contains clean BF16 master weights and has no quantization_config; it is not a packed W4A4 model. The source training checkpoint did not persist activation-observer scales. Clean-BF16 analysis is supported, while exact W4A4 continuation requires recalibration.

Intended use

These checkpoints are research artifacts for comparing approximately step-matched W4A4-QAT and BF16 training trajectories. They have not been evaluated here as general-purpose production models.

Validation

The export was checked for:

required Hugging Face model and tokenizer metadata;
readable safetensors headers and complete shard index;
exact index-to-shard key consistency;
BF16 tensor dtype throughout;
exact key and tensor-shape match against the native Qwen3-MoE architecture;
full source-artifact content comparison against the validated export;
Hugging Face path, byte-size, LFS SHA256, and metadata-download integrity.

SHA256SUMS covers every published file in this repository except the checksum manifest itself. Internal execution provenance is intentionally omitted from this public release.

Downloads last month: 111

Safetensors

Model size

31B params

Tensor type

BF16

Model tree for shawnzzzzz/Qwen3-30B-A3B-DAPO-FFN-W4A4-QAT-BF16Master-step-0060

Base model

Qwen/Qwen3-30B-A3B-Base

Finetuned

(64)

this model

Collection including shawnzzzzz/Qwen3-30B-A3B-DAPO-FFN-W4A4-QAT-BF16Master-step-0060

Qwen3-30B-A3B W4A4-QAT vs BF16 Checkpoints

Collection

Seven step-matched Qwen3-30B-A3B DAPO pairs: FFN W4A4-QAT BF16 master weights (not packed W4A4) and a BF16 baseline. • 14 items • Updated Jul 12