README.md

---
base_model:
- deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
datasets:
- open-thoughts/OpenThoughts2-1M
- Vinnnf/Hybrid-OpenThoughts2-1M-1.5B
library_name: transformers
license: apache-2.0
pipeline_tag: text-generation
---

# Thinkless: LLM Learns When to Think

![image/png](https://cdn-uploads.huggingface.co/production/uploads/646a1939c37ca1e12308fe81/SRxJKkSuC0y-oMB7SFeR6.png)

<table>
  <thead>
  </thead>
  <tbody>
    <tr>
      <td>📄 <strong>Paper Link</strong></td>
      <td><a href="https://arxiv.org/abs/2505.13379">ArXiv</a></td>
    </tr>
    <tr>
      <td>💻 <strong>RL Code</strong></td>
      <td><a href="https://github.com/VainF/Thinkless">VainF/Thinkless</a></td>
    </tr>
    <tr>
      <td>💻 <strong>SFT Code</strong></td>
      <td><a href="https://github.com/VainF/Reasoning-SFT">VainF/Reasoning-SFT</a></td>
    </tr>
    <tr>
      <td>🤖 <strong>RL Model</strong></td>
      <td><a href="https://huggingface.co/Vinnnf/Thinkless-1.5B-RL-DeepScaleR">Thinkless-1.5B-RL-DeepScaleR</a></td>
    </tr>
    <tr>
      <td>🐣 <strong>Warmup Model</strong></td>
      <td><a href="https://huggingface.co/Vinnnf/Thinkless-1.5B-Warmup">Thinkless-1.5B-Warmup</a></td>
    </tr>
    <tr>
      <td>📊 <strong>Data for Warmup</strong></td>
      <td><a href="https://huggingface.co/datasets/Vinnnf/Hybrid-OpenThoughts2-1M-1.5B">Hybrid-OpenThoughts2-1M-1.5B</a></td>
    </tr>
    <tr>
      <td>📊 <strong>Data for RL</strong></td>
      <td><a href="https://huggingface.co/datasets/agentica-org/DeepScaleR-Preview-Dataset">agentica-org/DeepScaleR-Preview-Dataset</a></td>
    </tr>
    <tr>
      <td> 🌐 <strong>Project Page</strong></td>
      <td><a href="https://sites.google.com/view/eagle-llm">Thinkless Website</a></td>
    </tr>
  </tbody>
</table>

## Introduction

> [!NOTE] 
> ***Can LLMs learn when to think?***

We propose Thinkless, a learnable framework that empowers an LLM to adaptively select between short-form and long-form reasoning based on both task complexity and the model's ability. Thinkless is trained under a reinforcement learning paradigm and employs two control tokens, \<short\> for concise responses and \<think\> for detailed reasoning. At the core of our method is a Decoupled Group Relative Policy Optimization (DeGRPO) algorithm, which decomposes the learning objective of hybrid reasoning into two components: (1) a control token loss that governs the selection of the reasoning mode, and (2) a response loss that improves the accuracy of the generated answers. This decoupled formulation enables fine-grained control over the contributions of each objective, stabilizing training and effectively preventing collapse observed in vanilla GRPO. Empirically, on several benchmarks such as Minerva Algebra, MATH-500, and GSM8K, Thinkless is able to reduce the usage of long-chain thinking by 50% - 90%, significantly reducing the computational cost of Reasoning Language Models.


## Pipeline

![image/png](https://cdn-uploads.huggingface.co/production/uploads/646a1939c37ca1e12308fe81/3mx8EJUyOvCtxPnYTcwbS.png)

## QuickStart
```python
from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "Vinnnf/Thinkless-1.5B-Warmup"

model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype="auto",
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)

instruction = "Please reason step by step, and put your final answer within \\boxed{}."
prompt = f"{instruction}
The arithmetic mean of 7, 2, $x$ and 10 is 9. What is the value of $x$?"

messages = [
    {"role": "user", "content": prompt}
]

text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True
)

think_mode = True
if think_mode:
    text = f"{text}<think>"
else:
    text = f"{text}<short>"

model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

generated_ids = model.generate(
    **model_inputs,
    max_new_tokens=4096
)
generated_ids = [
    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]
num_tokens = len(generated_ids[0])

response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]

print(text+response)
print(f"
Think Mode: {think_mode}")
print(f"Number of tokens: {num_tokens}")
```


## Citation
If you find this work helpful, please cite:
```
@article{fang2025thinkless,
  title={Thinkless: LLM Learns When to Think},
  author={Fang, Gongfan and Ma, Xinyin and Wang, Xinchao},
  journal={arXiv preprint arXiv:2505.13379},
  year={2025}
}

```