Instructions to use infly/INFIndo-Qwen3-32B-Preview with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use infly/INFIndo-Qwen3-32B-Preview with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="infly/INFIndo-Qwen3-32B-Preview")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("infly/INFIndo-Qwen3-32B-Preview")
model = AutoModelForCausalLM.from_pretrained("infly/INFIndo-Qwen3-32B-Preview")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use infly/INFIndo-Qwen3-32B-Preview with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "infly/INFIndo-Qwen3-32B-Preview"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "infly/INFIndo-Qwen3-32B-Preview",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/infly/INFIndo-Qwen3-32B-Preview

SGLang

How to use infly/INFIndo-Qwen3-32B-Preview with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "infly/INFIndo-Qwen3-32B-Preview" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "infly/INFIndo-Qwen3-32B-Preview",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "infly/INFIndo-Qwen3-32B-Preview" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "infly/INFIndo-Qwen3-32B-Preview",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use infly/INFIndo-Qwen3-32B-Preview with Docker Model Runner:
```
docker model run hf.co/infly/INFIndo-Qwen3-32B-Preview
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

INFIndo-Qwen3-32B-Preview

Model Overview

INFIndo-Qwen3-32B-Preview enhances the Indonesian language capabilities based on Qwen3-32B, which is produced through our designed data production pipeline using proprietary Indonesian language corpus. As of May 25th, 2025, this model has achieved the top performance among open-source models on the indonesian task in SEA-HELM (SouthEast Asian Holistic Evaluation of Language Models), which confirms the effectiveness of the improvement in the model's Indonesian language capabilities, and further validates the efficacy of our continual fine-tuning pipeline.

Evaluation Results

Model	SEA-HELM(ID)
INFIndo-Qwen3-32B-Preview	72.8
Qwen3-32B	71.9
DeepSeek-R1	72.1
Llama 3.3	70.4

Contributors

Supervisors

Wei Chu • Yuan Qi

Instruction Following Team

Tianchu Yao • Xiaoyu Tan • Dakuan Lu

Citation

If you find our model useful, please consider citing:

@misc{INFIndo_Preview,
  author       = {Yao, Tianchu and Tan, Xiaoyu and Lu, Dakuan and Chu, Wei and Qi, Yuan},
  title        = {INFIndo-Qwen3-32B-Preview},
  year         = {2025},
  month        = {May},
  howpublished = {Hugging Face},
  url          = {https://huggingface.co/infly/INFIndo-Qwen3-32B-Preview},
}

Downloads last month: 8

Safetensors

Model size

33B params

Tensor type

BF16

Model tree for infly/INFIndo-Qwen3-32B-Preview

Base model

Qwen/Qwen3-32B

Finetuned

(497)

this model

Quantizations

2 models