Instructions to use glorgao/PALU-1.5B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use glorgao/PALU-1.5B with Transformers:

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

pipe = pipeline("text-generation", model="glorgao/PALU-1.5B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("glorgao/PALU-1.5B")
model = AutoModelForCausalLM.from_pretrained("glorgao/PALU-1.5B")
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]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use glorgao/PALU-1.5B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "glorgao/PALU-1.5B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "glorgao/PALU-1.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/glorgao/PALU-1.5B

SGLang

How to use glorgao/PALU-1.5B 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 "glorgao/PALU-1.5B" \
    --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": "glorgao/PALU-1.5B",
		"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 "glorgao/PALU-1.5B" \
        --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": "glorgao/PALU-1.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use glorgao/PALU-1.5B with Docker Model Runner:
```
docker model run hf.co/glorgao/PALU-1.5B
```

Concise Reasoning in the Lens of Lagrangian Optimization

This model is trained by PALU, a post-training strategy for concise reasoning LLMs. PALU aims to enhance reasoning efficiency by optimizing the trade-off between accuracy and output verbosity.

Paper: https://arxiv.org/abs/2510.10168

Authors: Chengqian Gao, Haonan Li, Taylor W. Killian, Jianshu She, Renxi Wang, Liqun Ma, Zhoujun Cheng, Shibo Hao, and Zhiqiang Xu.

Key Features:

Concise reasoning: 65% fewer output tokens across five math benchmarks, with full reasoning preserved.
Performance: 15% higher accuracy than alternative solutions.
Efficiency: 20% faster compared with the GRPO training.

License: This repository and the model weights are released under the MIT License. The base model, DeepSeek-R1-Distill-Qwen-1.5B, is also under the MIT License, while its parent models, the Qwen-2.5 series, are licensed under the Apache 2.0 License.

Citation:

@article{gao2025concise,
  title={Concise Reasoning in the Lens of Lagrangian Optimization},
  author={Gao, Chengqian and Li, Haonan and Killian, Taylor W and She, Jianshu and Wang, Renxi and Ma, Liqun and Cheng, Zhoujun and Hao, Shibo and Xu, Zhiqiang},
  journal={arXiv preprint arXiv:2510.10168},
  year={2025}
}

Downloads last month: 5

Safetensors

Model size

2B params

Tensor type

BF16

Model tree for glorgao/PALU-1.5B

Base model

deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B

Finetuned

(634)

this model

Datasets used to train glorgao/PALU-1.5B

Paper for glorgao/PALU-1.5B

Concise Reasoning in the Lens of Lagrangian Optimization

Paper • 2510.10168 • Published Oct 11, 2025