---
license: apache-2.0
language:
- en
library_name: transformers
pipeline_tag: text-generation
base_model:
- Qwen/Qwen3-14B
base_model_relation: finetune
datasets:
- TheFinAI/FinCoT
tags:
- finance
- financial-reasoning
- reasoning
- grpo
- thefinai
- fino1
extra_gated_heading: Request access to Fin-o1-14B
extra_gated_description: Fin-o1-14B is released by The Fin AI for research. Access is granted automatically after you complete this short form.
extra_gated_button_content: Agree and access
extra_gated_prompt: This model is released for research purposes. It must not be used to provide personalized investment advice or to make automated financial decisions without human oversight. Use is also subject to the license of the base model (Qwen3, Apache 2.0).
extra_gated_fields:
  Full name: text
  Affiliation: text
  Country: country
  Intended use:
    type: select
    options:
    - Research
    - Education
    - Commercial evaluation
    - Other
  I agree to the terms above and will cite the Fino1 paper in any resulting publication: checkbox
---

# Fin-o1-14B

📄 [Paper](https://arxiv.org/abs/2502.08127) · 🤗 [Collection](https://huggingface.co/collections/TheFinAI/fin-o1-682702bbd25a7ef4512df24d) · 💻 [Code](https://github.com/the-finai/fino1) · 🏆 [Leaderboard](https://huggingface.co/spaces/TheFinAI/open-finllm-reasoning-leaderboard) · 🌐 [The Fin AI](https://thefin.ai)

**Fin-o1-14B** is a financial reasoning model fine-tuned from [Qwen3-14B](https://huggingface.co/Qwen/Qwen3-14B) with supervised fine-tuning on financial chain-of-thought data followed by GRPO reinforcement learning. It was introduced in *Fino1: On the Transferability of Reasoning-Enhanced LLMs and Reinforcement Learning to Finance* ([arXiv:2502.08127](https://arxiv.org/abs/2502.08127)).

> [!NOTE]
> **Fin-o1 vs. Fino1.** The two families come from the same paper but are different generations:
> | | Fino1 (2025-02/03) | Fin-o1 (2025-05) |
> |---|---|---|
> | 8B | [Fino1-8B](https://huggingface.co/TheFinAI/Fino1-8B) — Llama-3.1-8B-Instruct | [Fin-o1-8B](https://huggingface.co/TheFinAI/Fin-o1-8B) — Qwen3-8B |
> | 14B | [Fino1-14B](https://huggingface.co/TheFinAI/Fino1-14B) — Qwen2.5-14B-Instruct | **Fin-o1-14B** — Qwen3-14B |
> | Training | SFT + RL on GPT-4o reasoning paths | SFT + GRPO |
>
> For new work we recommend the **Fin-o1** models.

## Model Details

| | |
|---|---|
| Base model | [Qwen/Qwen3-14B](https://huggingface.co/Qwen/Qwen3-14B) |
| Architecture | Qwen3ForCausalLM, 40 layers, hidden size 5120 |
| Precision | bfloat16 |
| Context length | 40,960 tokens (`max_position_embeddings`) |
| Training data | [TheFinAI/FinCoT](https://huggingface.co/datasets/TheFinAI/FinCoT) |
| Training method | SFT, then GRPO |
| Language | English |
| License | Qwen3, Apache 2.0 (inherited from the base model) |

## Quick Start

```python
from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "TheFinAI/Fin-o1-14B"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype="auto", device_map="auto")

messages = [{
    "role": "user",
    "content": "A company's revenue grew from $120M to $150M while operating costs rose from $90M to $105M. "
               "By how many percentage points did the operating margin change?",
}]
inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt", return_dict=True).to(model.device)
output = model.generate(**inputs, max_new_tokens=1024)
print(tokenizer.decode(output[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))
```

## Evaluation

Fino1 and Fin-o1 are evaluated on FinQA, DocMath-Eval (simple / complex long), XBRL-Math and related financial reasoning tasks. See the paper and the [Open FinLLM Reasoning Leaderboard](https://huggingface.co/spaces/TheFinAI/open-finllm-reasoning-leaderboard) for results.

## Intended Use & Limitations

- **Intended use:** research on financial numerical reasoning, question answering over financial text and tables, and as a baseline for financial reasoning models.
- **Not intended for:** investment advice, trading decisions, or any automated financial decision without human review.
- The model can produce incorrect calculations or hallucinated figures, especially on long documents with multiple tables.
- Trained and evaluated on English data only.

## Citation

```bibtex
@misc{qian2025fino1transferabilityreasoningenhancedllms,
      title={Fino1: On the Transferability of Reasoning-Enhanced LLMs and Reinforcement Learning to Finance},
      author={Lingfei Qian and Weipeng Zhou and Yan Wang and Xueqing Peng and Han Yi and Yilun Zhao and Jimin Huang and Qianqian Xie and Jian-yun Nie},
      year={2025},
      eprint={2502.08127},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2502.08127},
}
```

## Contact

Questions and issues: open a discussion on this repository or on [GitHub](https://github.com/the-finai/fino1).