Instructions to use PKSGIN/qwen3-30b-selective-quant-mlx with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use PKSGIN/qwen3-30b-selective-quant-mlx with Transformers:

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

pipe = pipeline("text-generation", model="PKSGIN/qwen3-30b-selective-quant-mlx")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("PKSGIN/qwen3-30b-selective-quant-mlx")
model = AutoModelForCausalLM.from_pretrained("PKSGIN/qwen3-30b-selective-quant-mlx")
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 PKSGIN/qwen3-30b-selective-quant-mlx with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "PKSGIN/qwen3-30b-selective-quant-mlx"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "PKSGIN/qwen3-30b-selective-quant-mlx",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/PKSGIN/qwen3-30b-selective-quant-mlx

SGLang

How to use PKSGIN/qwen3-30b-selective-quant-mlx 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 "PKSGIN/qwen3-30b-selective-quant-mlx" \
    --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": "PKSGIN/qwen3-30b-selective-quant-mlx",
		"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 "PKSGIN/qwen3-30b-selective-quant-mlx" \
        --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": "PKSGIN/qwen3-30b-selective-quant-mlx",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use PKSGIN/qwen3-30b-selective-quant-mlx with Docker Model Runner:
```
docker model run hf.co/PKSGIN/qwen3-30b-selective-quant-mlx
```

This is the MLX Model that is final that can work on you MAC Book.

My laptop is M2 2023 with 32 GB Unified Memeory

Qwen3-30B-A3B Selective Quantization for Security & Code Analysis

A selectively quantized version of Qwen3-30B-A3B optimized for local inference on Apple Silicon, with preserved precision on code analysis experts.

MLX Model Here :

PKSGIN/qwen3-30b-selective-quant-mlx (Deprecated) PKSGIN/qwen3-30b-selective-quant-MixedMPW-mlx (NEW)

Model Description

This model applies heterogeneous quantization to Qwen3-30B-A3B's Mixture-of-Experts architecture:

9 coding experts → Q8 (8-bit precision)
119 general experts → Q4 (4-bit precision)
Router, attention, embeddings → FP16 (full precision)

The result: 18 GB on disk, runs at 52 tokens/sec on MacBook Pro M4 (32GB), with materially better code analysis quality than uniform Q4.

Why Selective Quantization?

Standard quantization tools (mlx_lm.convert, llama.cpp) compress all experts uniformly. For security and coding workloads, this degrades the exact experts you need most.

This model was profiled on 200 coding/security prompts to identify which experts activate for:

Infrastructure as Code (Terraform, Kubernetes)
Code review and vulnerability analysis
Security alert triage

Those 9 experts are preserved at Q8 while compressing the rest to Q4.

Target Use Cases

Local inference for sensitive workloads:

Infrastructure as Code review without cloud API calls
Code security analysis in air-gapped environments
SIEM alert triage with internal IP/hostname context
Pull request review for injection vulnerabilities

Zero data exfiltration. Zero tokens leaving your device.

Hardware Requirements

MacBook Pro M4 with 32GB unified memory (tested)
MacBook Pro M3/M2 with 32GB+ will work
Peak memory: ~18 GB during inference
Inference speed: 52 tokens/sec (M4 Pro)

Usage

# Install MLX
pip install mlx mlx-lm

# Run inference
python -m mlx_lm.generate \
    --model PKSGIN/qwen3-30b-selective-quant \
    --prompt "Review this Terraform for security issues: ..."

Performance

Metric	Value
Model size on disk	18 GB
Peak memory	18.1 GB
Tokens/sec (M4 Pro)	52
Prompt processing	4.4 tokens/sec
Compression ratio	3.0x vs FP16

Quantization Details

Profiled experts (coding ratio > 0.65):

[21, 27, 31, 43, 59, 66, 71, 113, 126]

Memory breakdown:

FP16 components (router/embeddings): 1.2 GB
Q8 coding experts: 2.1 GB
Q4 other experts: 16.7 GB
Total: 20 GB weights → 18 GB after MLX conversion

Limitations

Apple Silicon only — requires MLX framework
English-centric profiling — coding experts identified from English prompts
Domain-specific — optimized for code/security, may underperform on creative writing
Not benchmarked — no formal comparison vs uniform Q4 on CyberSecEval/HumanEval yet

Training & Profiling

Base model: Qwen/Qwen3-30B-A3B
Profiling: 200 coding/security prompts + 200 general prompts on A100 80GB
Quantization: Selective Q8/Q4 applied via custom pipeline
Conversion: Dequantized → MLX switch_mlp format → final Q4

Full pipeline: 4 scripts, ~$2 GPU rental, 72 minutes total.

Citation

@misc{qwen3-selective-quant-2026,
  author = {Prasanna Kanagasabai},
  title = {Qwen3-30B-A3B Selective Quantization for Security Workloads},
  year = {2026},
  publisher = {HuggingFace},
  url = {https://huggingface.co/PKSGIN/qwen3-30b-selective-quant}
}

License

Same as base model: Qwen License

Contact

Built by PK (@PKSGIN) — CISO, Singapore. Email : prasanna.in@gmail.com

Downloads last month: 4

Safetensors

Model size

5B params

Tensor type

U32

F16