This is a fork of https://huggingface.co/z-lab/Qwen3.6-27B-DFlash This model is still under training, and inference engine support may not be fully available yet due to architectural changes, including causal SWA layers. DFlash is a novel speculative decoding method that utilizes a lightweight block diffusion model for drafting. It enables efficient, high-quality parallel drafting that pushes the limits of inference speed. This model is the drafter component. It must be used in conjunction with the target model Qwen/Qwen3.6-27B. vLLM (We temporarily modify the installation through this PR to support interleaved SWA and ensure correct handling of target hidden states for optimal performance): Special thanks to David Wang for his outstanding engineering support on this project. We are also grateful to Modal, InnoMatrix, and Yotta Labs for providing the compute resources used to train this draft model. If you find DFlash useful, please cite our work. To share feedback on DFlash or request new model support, please fill out this form: DFlash Feedback.
Модальности:
Генерация текста
Задача: Генерация текста
Автор: QuixiAI
Теги: qwen3, feature-extraction, dflash, speculative-decoding, diffusion, efficiency, flash-decoding, qwen
Лайков: 3 | Загрузок: 30
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
