Qwen3-8B-INT4

Qwen3-8B quantized with torchao int4 weight only quantization, using hqq algorithm for improved accuracy, by PyTorch team. Use it directly or serve using vLLM for 62% VRAM reduction (6.27 GB needed) and 1.2x speedup on A100 GPUs.

Inference with vLLM

Install vllm nightly and torchao nightly to get some recent changes:

pip install vllm --pre --extra-index-url https://wheels.vllm.ai/nightly
pip install --pre torchao torch --index-url https://download.pytorch.org/whl/nightly/cu126

Serving

Then we can serve with the following command:

# Server
export MODEL=pytorch/Qwen3-8B-INT4
VLLM_DISABLE_COMPILE_CACHE=1 vllm serve $MODEL --tokenizer $MODEL -O3

# Client
curl http://localhost:8000/v1/chat/completions -H "Content-Type: application/json" -d '{
  "model": "pytorch/Qwen3-8B-INT4",
  "messages": [
    {"role": "user", "content": "Give me a short introduction to large language models."}
  ],
  "temperature": 0.6,
  "top_p": 0.95,
  "top_k": 20,
  "max_tokens": 32768
}'

Note: please use VLLM_DISABLE_COMPILE_CACHE=1 to disable compile cache when running this code, e.g. VLLM_DISABLE_COMPILE_CACHE=1 python example.py, since there are some issues with the composability of compile in vLLM and torchao, this is expected be resolved in pytorch 2.8.

Inference with Transformers

Install the required packages:

pip install git+https://github.com/huggingface/transformers@main
pip install --pre torchao torch --index-url https://download.pytorch.org/whl/nightly/cu126
pip install accelerate

Example:

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "pytorch/Qwen3-8B-INT4"

# load the tokenizer and the model
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype="auto",
    device_map="auto"
)

# prepare the model input
prompt = "Give me a short introduction to large language model."
messages = [
    {"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True,
    enable_thinking=True # Switches between thinking and non-thinking modes. Default is True.
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

# conduct text completion
generated_ids = model.generate(
    **model_inputs,
    max_new_tokens=32768
)
output_ids = generated_ids[0][len(model_inputs.input_ids[0]):].tolist() 

# parsing thinking content
try:
    # rindex finding 151668 (</think>)
    index = len(output_ids) - output_ids[::-1].index(151668)
except ValueError:
    index = 0

thinking_content = tokenizer.decode(output_ids[:index], skip_special_tokens=True).strip("\n")
content = tokenizer.decode(output_ids[index:], skip_special_tokens=True).strip("\n")

print("thinking content:", thinking_content)
print("content:", content)

Quantization Recipe

Install the required packages:

pip install git+https://github.com/huggingface/transformers@main
pip install --pre torchao torch --index-url https://download.pytorch.org/whl/nightly/cu126
pip install accelerate

Use the following code to get the quantized model:

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, TorchAoConfig

model_id = "Qwen/Qwen3-8B"

from torchao.quantization import Int4WeightOnlyConfig
quant_config = Int4WeightOnlyConfig(group_size=128, int4_packing_format="tile_packed_to_4d", int4_choose_qparams_algorithm="hqq")
# or for H100
# quant_config = Int4WeightOnlyConfig(group_size=128)
quantization_config = TorchAoConfig(quant_type=quant_config)
quantized_model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", torch_dtype=torch.bfloat16, quantization_config=quantization_config)
tokenizer = AutoTokenizer.from_pretrained(model_id)

# Push to hub
USER_ID = "YOUR_USER_ID"
MODEL_NAME = model_id.split("/")[-1]
save_to = f"{USER_ID}/{MODEL_NAME}-INT4"
quantized_model.push_to_hub(save_to, safe_serialization=False)
tokenizer.push_to_hub(save_to)

# Manual Testing
prompt = "Hey, are you conscious? Can you talk to me?"
messages = [
    {
        "role": "system",
        "content": "",
    },
    {"role": "user", "content": prompt},
]
templated_prompt = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True,
)
print("Prompt:", prompt)
print("Templated prompt:", templated_prompt)
inputs = tokenizer(
    templated_prompt,
    return_tensors="pt",
).to("cuda")
generated_ids = quantized_model.generate(**inputs, max_new_tokens=128)
output_text = tokenizer.batch_decode(
    generated_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False
)
print("Response:", output_text[0][len(prompt):])

Note: to push_to_hub you need to run

pip install -U "huggingface_hub[cli]"
huggingface-cli login

and use a token with write access, from https://huggingface.co/settings/tokens

Model Quality

We rely on lm-evaluation-harness to evaluate the quality of the quantized model.

<details> <summary> Reproduce Model Quality Results </summary>

Need to install lm-eval from source: https://github.com/EleutherAI/lm-evaluation-harness#install

baseline

lm_eval --model hf --model_args pretrained=Qwen/Qwen3-8B --tasks mmlu --device cuda:0 --batch_size 8

int4 weight only quantization with hqq (INT4)

export MODEL=pytorch/Qwen3-8B-INT4
# or
# export MODEL=Qwen/Qwen3-8B
lm_eval --model hf --model_args pretrained=$MODEL --tasks mmlu --device cuda:0 --batch_size 8

</details>

Peak Memory Usage

Results

<details> <summary> Reproduce Peak Memory Usage Results </summary>

We can use the following code to get a sense of peak memory usage during inference:

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, TorchAoConfig

# use "Qwen/Qwen3-8B" or "pytorch/Qwen3-8B-INT4"
model_id = "pytorch/Qwen3-8B-INT4"
quantized_model = AutoModelForCausalLM.from_pretrained(model_id, device_map="cuda:0", torch_dtype=torch.bfloat16)
tokenizer = AutoTokenizer.from_pretrained(model_id)

torch.cuda.reset_peak_memory_stats()

prompt = "Hey, are you conscious? Can you talk to me?"
messages = [
    {
        "role": "system",
        "content": "",
    },
    {"role": "user", "content": prompt},
]
templated_prompt = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True,
)
print("Prompt:", prompt)
print("Templated prompt:", templated_prompt)
inputs = tokenizer(
    templated_prompt,
    return_tensors="pt",
).to("cuda")
generated_ids = quantized_model.generate(**inputs, max_new_tokens=128)
output_text = tokenizer.batch_decode(
    generated_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False
)
print("Response:", output_text[0][len(prompt):])

mem = torch.cuda.max_memory_reserved() / 1e9
print(f"Peak Memory Usage: {mem:.02f} GB")

</details>

Model Performance

Our INT4 model is only optimized for batch size 1, so expect some slowdown with larger batch sizes, we expect this to be used in local server deployment for single or a few users where the decode tokens per second will matters more than the time to first token.

Results

A100

H100

Int4 weight only is optimized for batch size 1 and short input and output token length, please stay tuned for models optimized for larger batch sizes or longer token length.

<details> <summary> Reproduce Model Performance Results </summary>

Setup

Get vllm source code:

git clone [email protected]:vllm-project/vllm.git

Install vllm

VLLM_USE_PRECOMPILED=1 pip install --editable .

Run the benchmarks under vllm root folder:

benchmark_latency

baseline

export MODEL=Qwen/Qwen3-8B
vllm bench latency --input-len 256 --output-len 256 --model $MODEL --batch-size 1

INT4

export MODEL=pytorch/Qwen3-8B-INT4
VLLM_DISABLE_COMPILE_CACHE=1 vllm bench latency --input-len 256 --output-len 256 --model $MODEL --batch-size 1

benchmark_serving

We benchmarked the throughput in a serving environment.

Download sharegpt dataset:

wget https://huggingface.co/datasets/anon8231489123/ShareGPT_Vicuna_unfiltered/resolve/main/ShareGPT_V3_unfiltered_cleaned_split.json

Other datasets can be found in: https://github.com/vllm-project/vllm/tree/main/benchmarks

Note: you can change the number of prompts to be benchmarked with --num-prompts argument for benchmark_serving script.

baseline

Server:

export MODEL=Qwen/Qwen3-8B
vllm serve $MODEL --tokenizer $MODEL -O3

Client:

export MODEL=Qwen/Qwen3-8B
python benchmarks/benchmark_serving.py --backend vllm --dataset-name sharegpt --tokenizer $MODEL --dataset-path ./ShareGPT_V3_unfiltered_cleaned_split.json --model $MODEL --num-prompts 1

INT4

Server:

export MODEL=pytorch/Qwen3-8B-INT4
VLLM_DISABLE_COMPILE_CACHE=1 vllm serve $MODEL --tokenizer $MODEL -O3 --pt-load-map-location cuda:0

Client:

export MODEL=pytorch/Qwen3-8B-INT4
python benchmarks/benchmark_serving.py --backend vllm --dataset-name sharegpt --tokenizer $MODEL --dataset-path ./ShareGPT_V3_unfiltered_cleaned_split.json --model $MODEL --num-prompts 1

</details>

Paper: TorchAO: PyTorch-Native Training-to-Serving Model Optimization

The model's quantization is powered by TorchAO, a framework presented in the paper TorchAO: PyTorch-Native Training-to-Serving Model Optimization.

Abstract: We present TorchAO, a PyTorch-native model optimization framework leveraging quantization and sparsity to provide an end-to-end, training-to-serving workflow for AI models. TorchAO supports a variety of popular model optimization techniques, including FP8 quantized training, quantization-aware training (QAT), post-training quantization (PTQ), and 2:4 sparsity, and leverages a novel tensor subclass abstraction to represent a variety of widely-used, backend agnostic low precision data types, including INT4, INT8, FP8, MXFP4, MXFP6, and MXFP8. TorchAO integrates closely with the broader ecosystem at each step of the model optimization pipeline, from pre-training (TorchTitan) to fine-tuning (TorchTune, Axolotl) to serving (HuggingFace, vLLM, SGLang, ExecuTorch), connecting an otherwise fragmented space in a single, unified workflow. TorchAO has enabled recent launches of the quantized Llama 3.2 1B/3B and LlamaGuard3-8B models and is open-source at this https URL .

Resources

• Official TorchAO GitHub Repository: https://github.com/pytorch/ao
• TorchAO Documentation: https://docs.pytorch.org/ao/stable/index.html

Disclaimer

PyTorch has not performed safety evaluations or red teamed the quantized models. Performance characteristics, outputs, and behaviors may differ from the original models. Users are solely responsible for selecting appropriate use cases, evaluating and mitigating for accuracy, safety, and fairness, ensuring security, and complying with all applicable laws and regulations.

Nothing contained in this Model Card should be interpreted as or deemed a restriction or modification to the licenses the models are released under, including any limitations of liability or disclaimers of warranties provided therein.