metadata

license: mit
base_model:
  - Qwen/Qwen2.5-3B-Instruct
library_name: transformers
language:
  - en
  - zh
  - fr
  - es
  - pt
  - de
  - it
  - ru
  - ja
  - ko
  - vi
  - th
  - ar
  - fa
  - he
  - tr
  - cs
  - pl
  - hi
  - bn
  - ur
  - id
  - ms
  - lo
  - my
  - ceb
  - km
  - tl
  - nl
tags:
  - trl
  - Reasoning
  - open-llm
  - synthetic-data
  - Deepseek-R1
  - Qwen2.5
  - fine-tune
  - unsloth
  - Conversational
  - Agentic

Kyro-n1: A powerful family of models made for reasoning

This model uses some features from AIDC-AI/Marco-o1 tokenizer and this model is a Qwen2.5-3B fine-tune.

Kyro-n1 is a lightweight and fast reasoning model based on Qwen/Qwen2.5-3B-Instruct. We have further increased the quality of reasoning in certain aspects such as maths and science, but in this version, our main goal was maths and reasoning in general conversations. We intend to expand on this in future models. The whole purpose of Kyro is so that almost every device can run a reasoning model no matter their compute. This is why we are releasing 3B, 7B and 14B variants to achieve this goal.

Model Details

Developed by: Spestly (Open-Neo) & Kazex (Open-Neo)
Type: Causal Language Models
Training Stage: Pretraining & Post-training
Architecture: transformers with RoPE, SwiGLU, RMSNorm, Attention QKV bias and tied word embeddings
Number of Parameters: 3.09B
Number of Paramaters (Non-Embedding): 2.77B
Number of Layers: 36
Number of Attention Heads (GQA): 16 for Q and 2 for KV
Context Length: Full 32,768 tokens and generation 8192 tokens

Model Downloads

Kyro-n1 Models

	Training Data	Params	Input modalities	Output modalities	Context length	Download Link
Kyro (text only)	A new mix of publicly available online data.	3B	Multilingual Text	Multilingual Text and code	128k	🤗 HuggingFace
		7B	Multilingual Text	Multilingual Text and code	128k	🤗 HuggingFace
		14B	Multilingual Text	Multilingual Text and code	128k	🤗 HuggingFace

Kyro-1 Models

Usage

The code of Kyro-n1 (Qwen2.5) has been in the latest Hugging face transformers and we advise you to use the latest version of transformers.

With transformers<4.37.0, you will encounter the following error:

KeyError: 'qwen2'

Quickstart

from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "open-neo/Kyro-n1-3B"
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype="auto",
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)
prompt = "What do you think about CRISPR and its effect on the future of humanity?"
messages = [
    {"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
generated_ids = model.generate(
    **model_inputs,
    max_new_tokens=2048
)
generated_ids = [
    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]

Citation

If you find our work helpful, feel free to give us a cite.

@misc{qwen2.5,
    title = {Qwen2.5: A Party of Foundation Models},
    url = {https://qwenlm.github.io/blog/qwen2.5/},
    author = {Qwen Team},
    month = {September},
    year = {2024}
}
@article{qwen2,
      title={Qwen2 Technical Report}, 
      author={An Yang and Baosong Yang and Binyuan Hui and Bo Zheng and Bowen Yu and Chang Zhou and Chengpeng Li and Chengyuan Li and Dayiheng Liu and Fei Huang and Guanting Dong and Haoran Wei and Huan Lin and Jialong Tang and Jialin Wang and Jian Yang and Jianhong Tu and Jianwei Zhang and Jianxin Ma and Jin Xu and Jingren Zhou and Jinze Bai and Jinzheng He and Junyang Lin and Kai Dang and Keming Lu and Keqin Chen and Kexin Yang and Mei Li and Mingfeng Xue and Na Ni and Pei Zhang and Peng Wang and Ru Peng and Rui Men and Ruize Gao and Runji Lin and Shijie Wang and Shuai Bai and Sinan Tan and Tianhang Zhu and Tianhao Li and Tianyu Liu and Wenbin Ge and Xiaodong Deng and Xiaohuan Zhou and Xingzhang Ren and Xinyu Zhang and Xipin Wei and Xuancheng Ren and Yang Fan and Yang Yao and Yichang Zhang and Yu Wan and Yunfei Chu and Yuqiong Liu and Zeyu Cui and Zhenru Zhang and Zhihao Fan},
      journal={arXiv preprint arXiv:2407.10671},
      year={2024}
}
@misc{kyro-n1,
      title={Kyro-n1: A powerful family of models made for reasoning}, 
      author={Aayan Mishra and Krish Thumar},
      howpublished={https://huggingface.co/collections/open-neo/kyro-n1-67ab2e7bbc76a9aab3030c21},
      year={2025}
}