List your company

Speculative Decoding on LWM 8K context

2.54SpeedupBudgetDraft
0.93841.35421.772.1858May 29, 2026

Evaluation results

MethodLinks
BudgetDraftDrafter=68M, KV budget (B)=256, Draft length (γ)=52026.052.54
BudgetDraftDrafter=68M, KV budget (B)=512, Draft length (γ)=52026.052.54
BudgetDraftDrafter=68M, KV budget (B)=1024, Draft length (γ)=52026.052.54
BudgetDraftDrafter=68M, KV budget (B)=2048, Draft length (γ)=52026.052.54
EAGLE-3Drafter=draft head, Draft length (γ)=52026.051.64
TriForceDrafter=68M+7B, Draft length (γ)=52026.051.21
SD (sparse/full)Drafter=68M, Draft length (γ)=52026.051.19
ARDrafter=–, Draft length (γ)=52026.051