deepsweet/Qwen3.8-27B-DFlash2-FP16

This model was converted to FP16 from z-lab/Qwen3.8-27B-DFlash2 BF16.

What is "DFlash 2"?

DFlash 2 is a block-diffusion drafter for speculative decoding. It predicts a whole block of tokens in a single pass and keeps the top candidates at every position. A lightweight selector then traces one coherent path through them. Two-tap dynamic convolutions in the backbone keep the draft from decaying toward the end of the block. Decoding is lossless: greedy output matches the target model exactly, and sampling preserves its distribution.

What is "FP16"?

"FP16" is M1/M2 Apple Silicon only optimization that leads to a very noticeable prompt processing boost. See "Metal FP32 Vs BF16 Vs FP16 benchmark" and jundot/omlx/pull/880 for details.

Use the original model if you have M3+ Apple Silicon.