~/ai-muninn
search
⌘K
blog
model anatomy
github
中
~ / blog
/
tag / dflash2
❯
grep -r "#dflash2" ~/blog
4 matches
date
read
title
2026-09-28
17m
NVFP4 Without FP4 Hardware: Qwen3.8-27B on Two RTX 2080 Tis at 151.7 tok/s
#fastllm
#nvfp4
#quantization
#2080-ti
2026-09-28
21m
Dual RTX 2080 Ti FastLLM: 16% Faster Decoding and No More Head-of-Line Blocking
#fastllm
#dflash2
#speculative-decoding
#2080-ti
2026-09-21
24m
Two Modded RTX 2080 Tis Hit 153.8 tok/s on Qwen3.8-27B With FastLLM + DFlash2
#fastllm
#dflash2
#speculative-decoding
#2080-ti
2026-08-23
20m
[Benchmark] Qwen3.8-27B hits 65 tok/s on one DGX Spark — 12.3 without speculative decoding
#dgx-spark
#gb10
#sm121
#nvfp4
← back to all posts