@Ex0byt: the different flavors of specdec, and why I'm trying produce a Qwen-3.6-27b EAGLE-3 drafter for ya'll
Summary
Discussion of different flavors of speculative decoding and an attempt to produce a Qwen-3.6-27b EAGLE-3 drafter for the community.
View Cached Full Text
Cached at: 05/18/26, 10:31 AM
the different flavors of specdec, and why I’m trying produce a Qwen-3.6-27b EAGLE-3 drafter for ya’ll https://t.co/ZZvr28p2gU
Similar Articles
@charles_irl: Speculation Is All You Need. In this blog post, we announce the co-release (w/ Z Lab) of six more state-of-the-art DFla…
Modal and Z Lab release six new DFlash speculative decoding draft models for Qwen 3.x, achieving over 1000 tokens per second on a B200 and arguing that speculative decoding is the most impactful inference optimization.
@Ex0byt: And... Ladies and Gentlemen: Qwen3.6-27B-PRISM-PRO-DQ - enjoy!
Release of Qwen3.6-27B-PRISM-PRO-DQ, a dynamically quantized GGUF version of Qwen3.6-27B with bias/propaganda removal, preserving native MTP draft head and vision tower, enabling lossless speculative decoding for faster inference.
DFlash and Spec V2 Decoding (14 minute read)
Z Lab, SGLang, and Modal release DFlash, a new speculative decoding model for Qwen 3.5 397B-A17B that uses block diffusion and KV injection to achieve over 4x throughput improvement over baseline and 1.5x over native MTP.
Qwen3.6-27B speculative decoding gets better on heavier quants
Qwen3.6-27B model shows improved speculative decoding performance when using heavier quantization levels, enhancing inference efficiency.
z-lab/Qwen3.6-27B-DFlash
This article introduces Qwen3.6-27B-DFlash, a specialized drafter model for DFlash, a novel speculative decoding method using block diffusion to accelerate inference speed. It provides installation instructions for vLLM and SGLang to enable parallel drafting with the target Qwen3.6-27B model.