---
library_name: transformers
tags:
- torchao
- code
- math
- chat
- conversational
language:
- multilingual
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen3-8B/blob/main/LICENSE
pipeline_tag: text-generation
base_model:
- Qwen/Qwen3-8B
---

[Qwen3-8B](https://huggingface.co/Qwen/Qwen3-8B) quantized with [torchao](https://huggingface.co/docs/transformers/main/en/quantization/torchao) int4 weight only quantization, using [hqq](https://mobiusml.github.io/hqq_blog/) algorithm for improved accuracy, by PyTorch team.
Use it directly or serve using [vLLM](https://docs.vllm.ai/en/latest/) for 62% VRAM reduction 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 torchao
```

## Serving
Then we can serve with the following command:
```Shell
# Server
export MODEL=pytorch/Qwen3-8B-int4wo-hqq
VLLM_DISABLE_COMPILE_CACHE=1 vllm serve $MODEL --tokenizer $MODEL -O3
```

```Shell
# Client
curl http://localhost:8000/v1/chat/completions -H "Content-Type: application/json" -d '{
  "model": "pytorch/Qwen3-8B-int4wo-hqq",
  "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:
```Shell
pip install git+https://github.com/huggingface/transformers@main
pip install torchao
pip install torch
pip install accelerate
```

Example:
```Py
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "pytorch/Qwen3-8B-int4wo-hqq"

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

Use the following code to get the quantized model:
```Py
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, TorchAoConfig

model_id = "Qwen/Qwen3-8B"

from torchao.quantization import Int4WeightOnlyConfig
quant_config = Int4WeightOnlyConfig(group_size=128, use_hqq=True)
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}-int4wo-hqq"
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
```Shell
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](https://github.com/EleutherAI/lm-evaluation-harness) to evaluate the quality of the quantized model.

| Benchmark                        |                |                           |
|----------------------------------|----------------|---------------------------|
|                                  | Qwen3-8B       | Qwen3-8B-int4wo           | 
| **General**                      |                |                           |
| mmlu                             | 73.04          | 70.4                      |
| mmlu_pro                         | 53.81          | 52.79                     |
| bbh                              | 79.33          | 74.92                     |
| **Multilingual**                 |                |                           |
| mgsm_en_cot_en                   | 39.6           | 33.2                      |   
| m_mmlu (avg)                     | 57.17          | 54.06                     |
| **Math**                         |                |                           |
| gpqa_main_zeroshot               | 35.71          | 32.14                     |
| gsm8k                            | 87.79          | 86.28                     |
| leaderboard_math_hard (v3)       | 53.7           | 46.83                     |
| **Overall**                      | 60.02          | 56.33                     |

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

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

## baseline
```Shell
lm_eval --model hf --model_args pretrained=microsoft/Phi-4-mini-instruct --tasks mmlu --device cuda:0 --batch_size 8
```

## int4 weight only quantization with hqq (int4wo-hqq)
```Shell
export MODEL=pytorch/Qwen3-8B-int4wo-hqq
# 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

| Benchmark        |                |                                |
|------------------|----------------|--------------------------------|
|                  | Qwen3-8B       | Qwen3-8B-int4wo-hqq            | 
| Peak Memory (GB) | 16.47          | 6.27 (62% reduction)           |



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

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

```Py
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, TorchAoConfig

# use "Qwen/Qwen3-8B" or "pytorch/Qwen3-8B-int4wo-hqq"
model_id = "pytorch/Qwen3-8B-int4wo-hqq"
quantized_model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", 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 int4wo 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 machine)
| Benchmark (Latency)              |                |                          |
|----------------------------------|----------------|--------------------------|
|                                  | Qwen3-8B       | Qwen3-8B-int4wo-hqq      | 
| latency (batch_size=1)           | 3.52s          | 2.84s (1.24x speedup)    |

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:
```Shell
git clone git@github.com:vllm-project/vllm.git
```

Install vllm
```
VLLM_USE_PRECOMPILED=1 pip install --editable .
```

Run the benchmarks under `vllm` root folder:

## benchmark_latency

### baseline
```Shell
export MODEL=Qwen/Qwen3-8B
python benchmarks/benchmark_latency.py --input-len 256 --output-len 256 --model $MODEL --batch-size 1
```

### int4wo-hqq
```Shell
export MODEL=pytorch/Qwen3-8B-int4wo-hqq
VLLM_DISABLE_COMPILE_CACHE=1 python benchmarks/benchmark_latency.py --input-len 256 --output-len 256 --model $MODEL --batch-size 1
```

## benchmark_serving

We benchmarked the throughput in a serving environment.

Download sharegpt dataset: 

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

Client:
```Shell
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
```

### int4wo-hqq
Server:
```Shell
export MODEL=pytorch/Qwen3-8B-int4wo-hqq
VLLM_DISABLE_COMPILE_CACHE=1 vllm serve $MODEL --tokenizer $MODEL -O3 --pt-load-map-location cuda:0
```

Client:
```Shell
export MODEL=pytorch/Qwen3-8B-int4wo-hqq
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>



# 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.