Thinking-Space/UltraData-IF-1.5B

<h1 align="center">UltraData-IF-1.5B</h1>

<div align="center" style="line-height: 1;"> <a href="https://arxiv.org/pdf/2609.04172"><img src="https://img.shields.io/badge/paper-A42C25?style=for-the-badge&amp;logo=arxiv&amp;logoColor=white" alt="Paper"></a>&nbsp;&nbsp; <a href="https://github.com/Thinking-Space/One-Shot-OPD"><img src="https://img.shields.io/badge/One--Shot--OPD-000000?style=for-the-badge&amp;logo=github&amp;logoColor=white" alt="GitHub"></a>&nbsp;&nbsp; <a href="https://huggingface.co/papers/2609.04172"><img src="https://img.shields.io/badge/HF--Paper-%23FFD14D?style=for-the-badge&amp;logo=huggingface&amp;logoColor=black" alt="Hugging Face Paper"></a>&nbsp;&nbsp; <a href="https://huggingface.co/collections/Thinking-Space/one-shot-opd"><img src="https://img.shields.io/badge/Collection-%23FFD14D?style=for-the-badge&amp;logo=huggingface&amp;logoColor=black" alt="Hugging Face Collection"></a> <a href="https://x.com/HBX_hbx/status/2095716853196747175"><img src="https://img.shields.io/badge/Twitter-%23000000.svg?style=for-the-badge&logo=x&logoColor=white" alt="X Thread"></a> </div>

UltraData-IF-1.5B is an RL-trained instruction-following model initialized from DeepSeek-R1-Distill-Qwen-1.5B and trained on the instruction-following subset of openbmb/UltraData-SFT-2605.

This model is associated with the paper: Rethinking On-Policy Distillation of Large Language Models II: One Training Example

Model Description

UltraData-IF-1.5B serves as the teacher model for the instruction-following experiments in our paper. It is paired with DeepSeek-R1-Distill-Qwen-1.5B to study on-policy distillation in the instruction-following domain. The two models share the same architecture, tokenizer, and chat template, allowing token-level distillation without cross-model-family mismatch.

Key characteristics

• —Role in the paper: instruction-following teacher for OPD
• —Training method: GRPO
• —Base model: DeepSeek-R1-Distill-Qwen-1.5B （R1-Distill-1.5B）
• —Training data: instruction-following subset of openbmb/UltraData-SFT-2605
• —Paired student in OPD: DeepSeek-R1-Distill-Qwen-1.5B
• —Primary domain: Instruction following
• —Architecture: Qwen2ForCausalLM
• —Parameters: 1.78B

• —Precision: bfloat16
• —Vocabulary size: 151,936
• —Layers / hidden size: 28 / 1536

• —Attention heads: 12 query / 2 key-value (GQA)
• —Context length: 131,072 positions in config.json; tokenizer model_max_length is 16,384

Training Details

Evaluation

Scores for this checkpoint as the instruction-following teacher, alongside the student it is distilled into. Multi-IF is averaged over its eight languages; IFBench follows the standard prompt-level and instruction-level, strict and loose breakdown.

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "Thinking-Space/UltraData-IF-1.5B"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype="auto",
    device_map="auto",
)

messages = [{"role": "user", "content": "Write a haiku about distillation. Use exactly three lines."}]
inputs = tokenizer.apply_chat_template(
    messages,
    add_generation_prompt=True,
    return_tensors="pt",
).to(model.device)

outputs = model.generate(inputs, max_new_tokens=2048)
print(tokenizer.decode(outputs[0][inputs.shape[-1]:], skip_special_tokens=True))

Citation

If you use this model, please consider citing:

@article{fu2026rethinking,
  title={Rethinking on-policy distillation of large language models ii: One training example},
  author={Fu, Zixuan and He, Bingxiang and Zuo, Yuxin and Huang, Haohuan and Zhang, Jinqian and Xiao, Ruhang and Qian, Cheng and Luo, Qinyu and Gao, Huan-ang and Wang, Yudong and others},
  journal={arXiv preprint arXiv:2609.04172},
  year={2026}
}