Model Architecture: Qwen/Qwen3.5-4B
- Input: Text / Image
- Output: Text
Model Optimizations:
- Weight quantization: FP8
- Activation quantization: FP8
- Model size: 8.0 GB (reduced from 9.3 GB in BF16)
Release Date: 2026-05-11
Version: 1.0
Model Developers: RedHatAI

This model is a quantized version of Qwen/Qwen3.5-4B. Evaluation results and reproduction steps are provided below.

Model Optimizations

This model was obtained by quantizing the weights and activations of Qwen/Qwen3.5-4B to FP8 data type, ready for inference with vLLM.

This optimization reduces the model weights from 9.3 GB to 8.0 GB on disk (~14% reduction). Activations are quantized dynamically at inference time using per-tensor scaling, requiring no calibration data.

Only the weights and activations of the linear operators within transformer blocks are quantized using LLM Compressor.

Deployment

Use with vLLM

Initialize vLLM server:

Multimodal (vision + text):

bash

vllm serve RedHatAI/Qwen3.5-4B-FP8-dynamic \
  --reasoning-parser qwen3 \
  --max-model-len 262144

Text-only (lower memory):

bash

vllm serve RedHatAI/Qwen3.5-4B-FP8-dynamic \
  --reasoning-parser qwen3 \
  --max-model-len 262144 \
  --language-model-only

Send requests to the server:

python

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 = "RedHatAI/Qwen3.5-4B-FP8-dynamic"

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 using data-free FP8 dynamic quantization, as presented in the code snippet below.

python

from compressed_tensors.utils import save_mtp_tensors_to_checkpoint
from llmcompressor import oneshot
from llmcompressor.modifiers.quantization import QuantizationModifier
from transformers import AutoProcessor, AutoTokenizer, Qwen3_5ForConditionalGeneration

MODEL_ID = "Qwen/Qwen3.5-4B"

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)

recipe = QuantizationModifier(
    targets="Linear",
    scheme="FP8_DYNAMIC",
    ignore=IGNORE_LAYERS,
)

oneshot(model=model, recipe=recipe)

model.save_pretrained("Qwen3.5-4B-FP8-dynamic", save_compressed=True)
processor.save_pretrained("Qwen3.5-4B-FP8-dynamic")
save_mtp_tensors_to_checkpoint(source_model=MODEL_ID, dest_dir="Qwen3.5-4B-FP8-dynamic")

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/Qwen3.5-4B	RedHatAI/Qwen3.5-4B-FP8-dynamic	Recovery
Instruction Following	GSM8k-Platinum (0-shot)	94.2%	94.5%	100.3%
MMLU-Pro (0-shot)	79.3%	79.1%	99.9%
IFEval — prompt strict (0-shot)	88.0%	88.4%	100.5%
IFEval — instruction strict (0-shot)	91.2%	91.6%	100.4%
Reasoning	Math 500 (0-shot)	84.6%	84.7%	100.2%
AIME 2025 (0-shot)	85.0%	85.0%	100.0%
GPQA Diamond (0-shot)	76.8%	76.3%	99.3%

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)

bash

lm_eval --model local-chat-completions \
  --tasks gsm8k_platinum_cot_llama \
  --model_args "model=RedHatAI/Qwen3.5-4B-FP8-dynamic,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)

bash

lm_eval --model local-chat-completions \
  --tasks mmlu_pro_chat \
  --model_args "model=RedHatAI/Qwen3.5-4B-FP8-dynamic,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)

bash

lm_eval --model local-chat-completions \
  --tasks ifeval \
  --model_args "model=RedHatAI/Qwen3.5-4B-FP8-dynamic,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)

bash

lighteval endpoint litellm \
  "model_name=hosted_vllm/RedHatAI/Qwen3.5-4B-FP8-dynamic,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)

bash

lighteval endpoint litellm \
  "model_name=hosted_vllm/RedHatAI/Qwen3.5-4B-FP8-dynamic,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)

bash

lighteval endpoint litellm \
  "model_name=hosted_vllm/RedHatAI/Qwen3.5-4B-FP8-dynamic,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