---
title: Gradience-RL-Math
canonical_url: "https://www.modelscope.cn/models/Tesslate/Gradience-RL-Math"
md_url: "https://www.modelscope.cn/models/Tesslate/Gradience-RL-Math.md"
repository: Tesslate/Gradience-RL-Math
last_updated: 2025-04-13
base_model:
  - smirki/Gradience-T1-3B-Merged-2000
base_model_relation: finetune
parameters: 478.9M
tensor_type:
  - F32
library_name:
  - transformer
  - safetensors
  - pytorch
frameworks:
  - pytorch
downloads: 461
stars: 0
tags:
  - generated_from_trainer
  - unsloth
  - trl
  - grpo
---

# Gradience-RL-Math

> Gradience-RL-Math - Tesslate 在 ModelScope 开源的模型。Model Card for Gradience-RL-Math

Tesslate/Gradience-RL-Math 是 ModelScope 魔搭社区上的 478.9M 参数机器学习模型，基于 smirki/Gradience-T1-3B-Merged-2000 构建。

- **Repository**: Tesslate/Gradience-RL-Math
- **Parameters**: 478.9M
- **Base model**: smirki/Gradience-T1-3B-Merged-2000
- **Tags**: generated_from_trainer, unsloth, trl, grpo
- **Downloads**: 461
- **Stars**: 0
- **Last updated**: 2025-04-13

Source: https://www.modelscope.cn/models/Tesslate/Gradience-RL-Math

---

# Model Card for Gradience-RL-Math

This model is a fine-tuned version of [smirki/Gradience-T1-3B-Merged-2000](https://huggingface.co/smirki/Gradience-T1-3B-Merged-2000).
It has been trained using [TRL](https://github.com/huggingface/trl).

## Quick start

```python
from transformers import pipeline

question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="Tesslate/Gradience-RL-Math", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```

## Training procedure

 


This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).

### Framework versions

- TRL: 0.15.2
- Transformers: 4.50.0.dev0
- Pytorch: 2.5.1
- Datasets: 3.3.2
- Tokenizers: 0.21.0

## Citations

Cite GRPO as:

```bibtex
@article{zhihong2024deepseekmath,
    title        = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
    author       = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
    year         = 2024,
    eprint       = {arXiv:2402.03300},
}

```

Cite TRL as:
    
```bibtex
@misc{vonwerra2022trl,
	title        = {{TRL: Transformer Reinforcement Learning}},
	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
	year         = 2020,
	journal      = {GitHub repository},
	publisher    = {GitHub},
	howpublished = {\url{https://github.com/huggingface/trl}}
}
```