Model Architecture: Qwen3ForCausalLM
- Input: Text
- Output: Text
Model Optimizations:
- Weight quantization: INT4
Intended Use Cases:
- Reasoning.
- Function calling.
- Subject matter experts via fine-tuning.
- Multilingual instruction following.
- Translation.
Out-of-scope: Use in any manner that violates applicable laws or regulations (including trade compliance laws).
Release Date: 05/05/2025
Version: 1.0
Model Developers: RedHat (Neural Magic)

Model Optimizations

This model was obtained by quantizing the weights of Qwen/Qwen3-4B-Instruct-2507 to INT4 data type. This optimization reduces the number of bits per parameter from 16 to 4, reducing the disk size and GPU memory requirements by approximately 75%.

Only the weights of the linear operators within transformers blocks are quantized. Weights are quantized using a symmetric per-group scheme, with group size 128. The GPTQ algorithm is applied for quantization, as implemented in the llm-compressor library.

Deployment

This model can be deployed efficiently using the vLLM backend, as shown in the example below.

python

from vllm import LLM, SamplingParams
from transformers import AutoTokenizer

model_id = "RedHatAI/Qwen3-4B-Instruct-2507.w4a16"
number_gpus = 1
sampling_params = SamplingParams(temperature=0.6, top_p=0.95, top_k=20, min_p=0, max_tokens=256)

messages = [
    {"role": "user", "content": prompt}
]

tokenizer = AutoTokenizer.from_pretrained(model_id)

messages = [{"role": "user", "content": "Give me a short introduction to large language model."}]

prompts = tokenizer.apply_chat_template(messages, add_generation_prompt=True, tokenize=False)

llm = LLM(model=model_id, tensor_parallel_size=number_gpus)

outputs = llm.generate(prompts, sampling_params)

generated_text = outputs[0].outputs[0].text
print(generated_text)

vLLM aslo supports OpenAI-compatible serving. See the documentation for more details.

Creation

Creation details

This model was created with [llm-compressor](https://github.com/vllm-project/llm-compressor) by running the code snippet below.

python

from llmcompressor.modifiers.quantization import GPTQModifier
from llmcompressor.transformers import oneshot
from transformers import AutoModelForCausalLM, AutoTokenizer

# Load model
model_stub = "Qwen/Qwen3-4B-Instruct"
model_name = model_stub.split("/")[-1]

num_samples = 1024
max_seq_len = 8192

model = AutoModelForCausalLM.from_pretrained(model_stub)

tokenizer = AutoTokenizer.from_pretrained(model_stub)

def preprocess_fn(example):
  return {"text": tokenizer.apply_chat_template(example["messages"], add_generation_prompt=False, tokenize=False)}

ds = load_dataset("neuralmagic/LLM_compression_calibration", split="train")
ds = ds.map(preprocess_fn)

# Configure the quantization algorithm and scheme
recipe = GPTQModifier(
    ignore: ["lm_head"]
    config_groups={"group_0": {"targets": ["Linear"], "weights": { "num_bits": 4, "type": int, "strategy": "group", "group_size": 128, "symmetric": true, "dynamic": false, "observer": "mse" } } },
    dampening_frac=0.01,
)

# Apply quantization
oneshot(
    model=model,
    dataset=ds, 
    recipe=recipe,
    max_seq_length=max_seq_len,
    num_calibration_samples=num_samples,
)

# Save to disk in compressed-tensors format
save_path = model_name + "-quantized.w4a16"
model.save_pretrained(save_path)
tokenizer.save_pretrained(save_path)
print(f"Model and tokenizer saved to: {save_path}")

Evaluation

The model was evaluated on the ifeval, mmlu_pro and gsm8k_platinum using lm-evaluation-harness, on reasoning tasks using lighteval. vLLM was used for all evaluations.

Evaluation details

Deploy using vllm to create an OpenAI-compatible API endpoint:

vLLM:

vllm serve RedHatAI/Qwen3-4B-Instruct-2507.w4a16 --max-model-len 262144 --reasoning-parser deepseek_r1

lm-evaluation-harness

lm_eval --model local-chat-completions \
  --tasks mmlu_pro_chat \
  --model_args "model=RedHatAI/Qwen3-4B-Instruct-2507.w4a16,max_length=262144,base_url=http://0.0.0.0:8000/v1/chat/completions,num_concurrent=64,max_retries=3,tokenized_requests=False,tokenizer_backend=None,timeout=1200" \
  --num_fewshot 0 \
  --apply_chat_template \
  --gen_kwargs "do_sample=True,temperature=0.6,top_p=0.95,top_k=20,min_p=0.0,max_gen_toks=64000

lm_eval --model local-chat-completions \
  --tasks ifeval \
  --model_args "model=RedHatAI/Qwen3-4B-Instruct-2507.w4a16,max_length=262144,base_url=http://0.0.0.0:8000/v1/chat/completions,num_concurrent=64,max_retries=3,tokenized_requests=False,tokenizer_backend=None,timeout=1200" \
  --num_fewshot 0 \
  --apply_chat_template \
  --gen_kwargs "do_sample=True,temperature=0.6,top_p=0.95,top_k=20,min_p=0.0,max_gen_toks=64000

lm_eval --model local-chat-completions \
  --tasks gsm8k_platinum_cot_llama \
  --model_args "model=RedHatAI/Qwen3-4B-Instruct-2507.w4a16,max_length=262144,base_url=http://0.0.0.0:8000/v1/chat/completions,num_concurrent=64,max_retries=3,tokenized_requests=False,tokenizer_backend=None,timeout=1200" \
  --num_fewshot 0 \
  --apply_chat_template \
  --gen_kwargs "do_sample=True,temperature=0.6,top_p=0.95,top_k=20,min_p=0.0,max_gen_toks=64000

lighteval

lighteval_model_arguments.yaml

yaml

model_parameters:
  model_name: RedHatAI/Qwen3-4B-Instruct-2507.w4a16
  dtype: auto
  gpu_memory_utilization: 0.9
  max_model_length: 40960
  generation_parameters:
    temperature: 0.6
    top_k: 20
    min_p: 0.0
    top_p: 0.95
    max_new_tokens: 32000

lighteval endpoint litellm lighteval_model_arguments.yaml  \
  "aime25|0,math_500|0,gpqa:diamond|0"

Accuracy

Benchmark	Qwen3-4B Instruct	Qwen3-4B Instruct.w4a16 (this model)	Recovery (%)
GSM8k Platinum (5-shot)	95.62	96.14	100.55
MMLU-CoT (5-shot)	77.55	76.83	99.08
MMLU-Pro (5-shot)	70.13	68.94	98.31
IfEval	89.01	88.61	99.55
Math 500	84.33	84.20	99.85

RedHatAI/Qwen3-4B-Instruct-2507-quantized.w4a16

Qwen3-4B-Instruct-2507.w4a16

Model Overview

Model Optimizations

Deployment

Creation

Evaluation

Accuracy

RedHatAI/
Qwen3-4B-Instruct-2507-quantized.w4a16