I benchmark DFlash 2 (PR build) in llama.cpp on Qwen 3.8 27B against all speculative methods for 3 days. 2.26x on 100 real coding prompts, 4.68x with one n-gram drafter on top. Up to 8x on specific cases.

英文标题：I benchmark DFlash 2 (PR build) in llama.cpp on Qwen 3.8 27B against all speculative methods for 3 days. 2.26x on 100 real coding prompts, 4.68x with one n-gram drafter on top. Up to 8x on specific cases.

Hey guys, Inco AI shipped DFlash 2 a few days ago with a drafter for Qwen 3.8 27B and a llama.cpp PR. I built the PR and ran it against plain decoding, MTP, the n-gram lookup drafters, and my July DFlash 1 numbers on Qwen 3.6 27B for 3 days. One RTX PRO 6000, concurrency 1, about three days of