Inco AI Releases DFlash 2 Draft Model, Speeding Up GLM-5.3-Flash by 2.8x

Inco AI has released DFlash 2, a draft model for GLM-5.3-Flash on Hugging Face. Using block-diffusion based speculative decoding, it accelerates inference by 2.8x.

2026-08-28 ~ 2026-08-28 · 3 related posts

• DFlash 2 Introduces Block-Diffusion Speculative Decoding to Speed Up GLM-5.3-Flash — songhan_mit · 2026-08-28

2 near-duplicate retellings: Lianhuiq · TheZachMueller