z-lab

    z-lab/dflash

    DFlash: Block Diffusion for Flash Speculative Decoding

    backend
    Python
    MIT
    6.2K stars
    441 forks
    6.2K GitHub watchers
    Updated Oct 27, 2026
    View on GitHub

    Build with Backblaze B2

    SDKs, agent skills, IDE extensions, and reference pipelines from Backblaze Labs. All open source.

    Explore Backblaze Labs

    Loading star history...

    Where it stands

    Rank, growth pace and release cadence relative to everything GStars tracks.

    1 release in the last 90 days, the latest 70 days ago, with 4 contributors on GitHub.

    Peaked at #34 in the weekly ranking the week of Apr 19, its best in the past year.

    Use Cases & Benefits

    • Implements a lightweight block diffusion model for efficient speculative decoding in large language models.
    • Enables high-quality parallel drafting to accelerate inference while maintaining output quality.
    • Use for accelerating inference in large language models like Qwen and LLaMA with speculative decoding.
    • Use for integrating with vLLM, SGLang, or Transformers backends to optimize model serving pipelines.
    • Use for researchers or developers training custom DFlash draft models to speed up LLM decoding.

    Should you depend on it?

    License terms, project hygiene and risks, from GitHub metadata and GStars' snapshots.

    License

    MIT

    Permissive: commercial use, modification and redistribution are allowed with attribution.

    Community files

    • README: present

    Watch-outs

    No watch-outs from GStars' checks.

    General guidance from repository metadata, not legal advice. How we score

    From the README

    DFlash is a lightweight block diffusion model designed for speculative decoding. It enables efficient and high-quality parallel drafting.

    Excerpt from the z-lab/dflash README on GitHub (MIT license). © its authors.Read the full README on GitHub

    Daily snapshots

    Stars gained per day, and how forks and open issues moved, from GStars' daily snapshots

    Forks

    441+43

    Open issues

    106+21

    Compared with peers in Backend

    Fastest-growing repositories GStars tracks in this cohort for the week of Nov 1, by 7-day star growth

    z-lab/dflash compared with Backend repositories by stars, 7-day star growth, health score and last release
    RepositoryStars+7dHealthLast release
    anthropics/skills187.8K+2.9K15—
    DuarteSantos8/openGym19.7K+2.4K602 days ago
    shiyu-coder/Kronos47.0K+2.3K16—
    NationalSecurityAgency/ghidra98.2K+2.3K6414 days ago
    microsoft/VibeVoice56.0K+1.1K29—
    z-lab/dflash6.2K—2470 days ago
    Category median56.0K+2.3K29—

    Discover Repositories

    Search across tracked repositories by name or description