modal-labs / GLM-5.3-Flash-DFlash

GLM-5.3-Flash-DFlash

Paper | Github | Blog

This repository contains a DFlash draft model for zai-org/GLM-5.3-Flash . It is not a standalone language model. It is intended to be paired with the target model in a speculative decoding server.

DFlash uses a lightweight block diffusion draft model to propose multiple tokens in parallel. The target model verifies those proposals, improving serving throughput while preserving the target model's output distribution.

Quick Start

GLM-5.3-Flash needs the DFlash capture hooks in the glm5_next model, available on SGLang main. An example deployment is:

python -m sglang.launch_server \
  --model-path zai-org/GLM-5.3-Flash \
  --tp-size 4 \
  --trust-remote-code \
  --speculative-algorithm DFLASH \
  --speculative-draft-model-path modal-labs/GLM-5.3-Flash-DFlash \
  --speculative-dflash-block-size 8 \
  --speculative-draft-model-quantization unquant \
  --speculative-draft-attention-backend trtllm_mha \
  --speculative-draft-kv-cache-dtype fp8_e4m3 \
  --host 0.0.0.0 \
  --port 30000

Keep the draft model unquantized. Quantizing it lowers the accept length.

License

Distributed under the MIT License , inherited from the target model.

Citation

If you find DFlash useful, please cite the original paper:

@article{chen2026dflash,
  title   = {{DFlash: Block Diffusion for Flash Speculative Decoding}},
  author  = {Chen, Jian and Liang, Yesheng and Liu, Zhijian},
  journal = {arXiv preprint arXiv:2602.06036},
  year    = {2026}
}