DFlash: Block Diffusion for Flash Speculative Decoding
Paper
• 2602.06036 • Published
• 42
DFlash
Block Diffusion for Flash Speculative Decoding
-
-
z-lab/Qwen3-4B-DFlash-b16
Text Generation • 0.5B • Updated • 19.6k • 22 -
z-lab/Qwen3-8B-DFlash-b16
Text Generation • Updated • 7.12k • 20 -
z-lab/LLaMA3.1-8B-Instruct-DFlash-UltraChat
Text Generation • 1B • Updated • 1.49k • 2 -
z-lab/Qwen3-Coder-30B-A3B-DFlash
Text Generation • Updated • 1.19k • 27 -
z-lab/gpt-oss-20b-DFlash
Text Generation • 0.8B • Updated • 433 • 10