DFlash
How it works
1) A lightweight block-diffusion draft model generates a whole block of candidate draft tokens in a single forward pass instead of producing them sequentially. 2) The drafter is conditioned on context features extracted from the target model, improving draft accuracy. 3) The target model verifies the block of draft tokens in parallel. 4) Accepted tokens are kept and the process stays lossless — the target model output distribution is preserved.
Problem solved
Autoregressive LLMs decode sequentially, token by token, causing high inference latency and poor GPU utilization. Existing speculative decoding mitigates this, but the draft model itself still runs autoregressively and sequentially, which limits practical speedups.
Key mechanisms
Strengths & limitations
Components
A lightweight block-diffusion model that generates an entire block of draft tokens in a single forward pass rather than sequentially.
Official
A mechanism conditioning the drafter on context features extracted from the target model, increasing draft quality and the acceptance rate.
The target model (large LLM) that verifies the block of draft tokens in parallel and guarantees lossless generation.
Implementation
Evolution
Introduction of DFlash — speculative decoding with a lightweight block-diffusion drafter; accepted at ICML 2026, with public code.
Computational complexity
Per the originating paper, DFlash achieves over 6x lossless acceleration across a range of models and tasks and up to 2.5x higher speedup than the SOTA EAGLE-3 method. Per-model/per-task figures are in the paper (arXiv:2602.06036, ICML 2026).
Parallelism
Drafting generates a whole block of tokens in parallel in a single forward pass, and target-model verification runs in parallel over the entire block.
Hardware requirements
The method targets improved GPU utilization and reduced LLM inference latency; parallel drafting and verification exploit GPU accelerator parallelism.