Qwen3.5-9B-quantized.w4a16

Model Overview

• Model Architecture: Qwen/Qwen3.5-9B
  • Input: Text / Image
  • Output: Text
• Model Optimizations:
  • Weight quantization: INT4
  • Activation quantization: None
  • Model size: 11 GB (reduced from 19.3 GB in BF16)
• Release Date: 2026-04-24
• Version: 1.0
• Model Developers: RedHatAI

This model is a quantized version of Qwen/Qwen3.5-9B. Evaluation results and reproduction steps are provided below.

Model Optimizations

This model was obtained by quantizing the weights of Qwen/Qwen3.5-9B to INT4 data type while keeping activations in original precision, ready for inference with vLLM.

This optimization reduces the model weights from 19.3 GB to 11 GB on disk (~43% reduction). The reduction is less than the theoretical 75% because the vision encoder, token embeddings, and linear attention layers remain in BF16.

Only the weights of the linear operators within transformer blocks are quantized using LLM Compressor. The vision encoder, token embeddings, and linear attention layers are not quantized.

Deployment

Use with vLLM

1. Initialize vLLM server:

Multimodal (vision + text):

vllm serve inference-optimization/Qwen3.5-9B-quantized.w4a16 \
  --reasoning-parser qwen3 \
  --max-model-len 262144

Text-only (lower memory):

vllm serve inference-optimization/Qwen3.5-9B-quantized.w4a16 \
  --reasoning-parser qwen3 \
  --max-model-len 262144 \
  --language-model-only

2. Send requests to the server:

from openai import OpenAI

openai_api_key = "EMPTY"
openai_api_base = "http://localhost:8000/v1"

client = OpenAI(
    api_key=openai_api_key,
    base_url=openai_api_base,
)

model = "inference-optimization/Qwen3.5-9B-quantized.w4a16"

messages = [
    {"role": "user", "content": "Explain quantum mechanics clearly and concisely."},
]

outputs = client.chat.completions.create(
    model=model,
    messages=messages,
)

generated_text = outputs.choices[0].message.content
print(generated_text)

Creation

This model was created by applying LLM Compressor with calibration samples from Open-Platypus, as presented in the code snippet below.

from compressed_tensors.utils import save_mtp_tensors_to_checkpoint
from datasets import load_dataset
from llmcompressor import oneshot
from llmcompressor.modifiers.quantization import GPTQModifier
from transformers import AutoProcessor, AutoTokenizer, Qwen3_5ForConditionalGeneration

MODEL_ID = "Qwen/Qwen3.5-9B"
NUM_CALIBRATION_SAMPLES = 1024
MAX_SEQUENCE_LENGTH = 8192

IGNORE_LAYERS = [
    "re:.*lm_head",
    "re:.*embed_tokens$",
    "re:.*visual.*",
    "re:.*model.visual.*",
    "re:.*linear_attn.*",
]

model = Qwen3_5ForConditionalGeneration.from_pretrained(MODEL_ID, dtype="auto")
tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)
processor = AutoProcessor.from_pretrained(MODEL_ID)

ds = load_dataset("garage-bAInd/Open-Platypus", split=f"train[:{NUM_CALIBRATION_SAMPLES}]")
ds = ds.shuffle(seed=42)

def preprocess(ex):
    text = ex["instruction"]
    if ex.get("input"):
        text += "\n" + ex["input"]
    return {"text": text}

def tokenize(sample):
    return tokenizer(
        sample["text"],
        padding=False,
        max_length=MAX_SEQUENCE_LENGTH,
        truncation=True,
        add_special_tokens=False,
    )

ds = ds.map(preprocess).map(tokenize, remove_columns=ds.column_names)

recipe = GPTQModifier(
    targets="Linear",
    scheme="W4A16",
    sequential_targets=["Qwen3_5DecoderLayer"],
    ignore=IGNORE_LAYERS,
    dampening_frac=0.05,
)

oneshot(
    model=model,
    dataset=ds,
    recipe=recipe,
    max_seq_length=MAX_SEQUENCE_LENGTH,
    num_calibration_samples=NUM_CALIBRATION_SAMPLES,
)

model.save_pretrained("Qwen3.5-9B-quantized.w4a16", save_compressed=True)
processor.save_pretrained("Qwen3.5-9B-quantized.w4a16")
save_mtp_tensors_to_checkpoint(source_model=MODEL_ID, dest_dir="Qwen3.5-9B-quantized.w4a16")

Package versions

• llm-compressor==0.10.1.dev44+g437f8afe
• compressed-tensors==0.14.1a20260325
• transformers==5.3.0
• vllm==0.18.1
• lm-eval — neuralmagic/lm-evaluation-harness@741f1d8 (branch: mmlu-pro-chat-variant)
• lighteval — neuralmagic/lighteval@6f0f351 (branch: eldar-fix-litellm)

Evaluation

This model was evaluated on GSM8k-Platinum, MMLU-Pro, IFEval, Math 500, AIME 2025, and GPQA Diamond using lm-evaluation-harness and lighteval, with inference served via vLLM.

Accuracy

Category	Benchmark	Qwen/Qwen3.5-9B	inference-optimization/Qwen3.5-9B-quantized.w4a16	Recovery
Instruction Following	GSM8k-Platinum (0-shot)	94.4%	94.6%	100.1%
	MMLU-Pro (0-shot)	82.4%	81.7%	99.1%
	IFEval — prompt strict (0-shot)	89.5%	87.7%	97.9%
	IFEval — instruction strict (0-shot)	92.5%	91.3%	98.7%
Reasoning	Math 500 (0-shot)	85.2%	83.8%	98.4%
	AIME 2025 (0-shot)	85.4%	68.8%	80.5%
	GPQA Diamond (0-shot)	82.2%	77.9%	94.9%

Reproduction

The results were obtained using the following commands. GSM8k-Platinum, MMLU-Pro, IFEval, Math 500, and GPQA Diamond were each run 3 times with different seeds and results averaged. AIME 2025 was run 8 times. The vLLM server was started with --language-model-only for all evaluations.

GSM8k-Platinum (lm-eval, 0-shot, 3 repetitions)

lm_eval --model local-chat-completions \
  --tasks gsm8k_platinum_cot_llama \
  --model_args "model=inference-optimization/Qwen3.5-9B-quantized.w4a16,max_length=96000,base_url=http://0.0.0.0:8000/v1/chat/completions,num_concurrent=100,max_retries=3,tokenized_requests=False,tokenizer_backend=None,timeout=3600" \
  --num_fewshot 0 \
  --apply_chat_template \
  --output_path results_gsm8k_platinum.json \
  --seed <SEED> \
  --gen_kwargs "do_sample=True,temperature=1.0,top_p=0.95,top_k=20,min_p=0.0,presence_penalty=1.5,repetition_penalty=1.0,max_gen_toks=65536,seed=<SEED>"

Seeds used: 42, 1234, 4158

MMLU-Pro (lm-eval, 0-shot, 3 repetitions)

lm_eval --model local-chat-completions \
  --tasks mmlu_pro_chat \
  --model_args "model=inference-optimization/Qwen3.5-9B-quantized.w4a16,max_length=96000,base_url=http://0.0.0.0:8000/v1/chat/completions,num_concurrent=100,max_retries=3,tokenized_requests=False,tokenizer_backend=None,timeout=3600" \
  --num_fewshot 0 \
  --apply_chat_template \
  --output_path results_mmlu_pro.json \
  --seed <SEED> \
  --gen_kwargs "do_sample=True,temperature=1.0,top_p=0.95,top_k=20,min_p=0.0,presence_penalty=1.5,repetition_penalty=1.0,max_gen_toks=65536,seed=<SEED>"

Seeds used: 42, 1234, 4158

IFEval (lm-eval, 0-shot, 3 repetitions)

lm_eval --model local-chat-completions \
  --tasks ifeval \
  --model_args "model=inference-optimization/Qwen3.5-9B-quantized.w4a16,max_length=96000,base_url=http://0.0.0.0:8000/v1/chat/completions,num_concurrent=100,max_retries=3,tokenized_requests=False,tokenizer_backend=None,timeout=3600" \
  --num_fewshot 0 \
  --apply_chat_template \
  --output_path results_ifeval.json \
  --seed <SEED> \
  --gen_kwargs "do_sample=True,temperature=1.0,top_p=0.95,top_k=20,min_p=0.0,presence_penalty=1.5,repetition_penalty=1.0,max_gen_toks=65536,seed=<SEED>"

Seeds used: 42, 1234, 4158

Math 500 (lighteval, 0-shot, 3 repetitions)

lighteval endpoint litellm \
  "model_name=hosted_vllm/inference-optimization/Qwen3.5-9B-quantized.w4a16,provider=hosted_vllm,base_url=http://0.0.0.0:8000/v1,timeout=3600,concurrent_requests=100,generation_parameters={temperature:1.0,max_new_tokens:65536,top_p:0.95,top_k:20,min_p:0.0,presence_penalty:1.5,repetition_penalty:1.0,seed:<SEED>}" \
  "math_500@k=1@n=1|0" \
  --output-dir results_math500 \
  --save-details

Seeds used: 42, 1234, 4158

AIME 2025 (lighteval, 0-shot, 8 repetitions)

lighteval endpoint litellm \
  "model_name=hosted_vllm/inference-optimization/Qwen3.5-9B-quantized.w4a16,provider=hosted_vllm,base_url=http://0.0.0.0:8000/v1,timeout=3600,concurrent_requests=100,generation_parameters={temperature:1.0,max_new_tokens:65536,top_p:0.95,top_k:20,min_p:0.0,presence_penalty:1.5,repetition_penalty:1.0,seed:<SEED>}" \
  "aime25@k=1@n=1|0" \
  --output-dir results_aime25 \
  --save-details

Seeds used: 42, 1234, 1356, 3344, 4158, 5322, 5678, 9843

GPQA Diamond (lighteval, 0-shot, 3 repetitions)

lighteval endpoint litellm \
  "model_name=hosted_vllm/inference-optimization/Qwen3.5-9B-quantized.w4a16,provider=hosted_vllm,base_url=http://0.0.0.0:8000/v1,timeout=3600,concurrent_requests=100,generation_parameters={temperature:1.0,max_new_tokens:65536,top_p:0.95,top_k:20,min_p:0.0,presence_penalty:1.5,repetition_penalty:1.0,seed:<SEED>}" \
  "gpqa:diamond@k=1@n=1|0" \
  --output-dir results_gpqa_diamond \
  --save-details

Seeds used: 42, 1234, 4158