david3684/FMLM-B-LM1B

FMLM (LM1B)

For more information and implementation details, visit our github repository: FLM.

Model Details

The model has a context length of 128, distilled from 1M steps trained FLM checkpoint trained on the One Billion Words (LM1B) dataset.

For more details, please see our paper:

• arXiv:2602.16813
• Huggingface Papers

Citation

Please cite our work using the bibtex below:

BibTeX:

@article{lee2026flow,
    title={Flow Map Language Models: One-step Language Modeling via Continuous Denoising},
    author={Chanhyuk Lee and Jaehoon Yoo and Manan Agarwal
            and Sheel Shah and Jerry Huang
            and Aditi Raghunathan and Seunghoon Hong
            and Nicholas M. Boffi and Jinwoo Kim},
    journal={arXiv preprint arXiv:2602.16813},
    year={2026},
}

Model Card Contact

Chanhyuk Lee (chanhyuk-lee@kaist.ac.kr)