Unverified50% confidenceSolutionExact time
投机解码通过小型草稿模型提前生成候选token序列，可在保持输出质量前提下将LLM有效吞吐量提升2-3倍
1
Sources
50%
Confidence
Long-term
Relevance
7/22/2026
First Seen
Sources
Related Claims
VerifiedSpeculative Decoding uses a small, fast draft model to predict multiple tokens in advance, which are then verified in parallel all at once by the large model.79% similarUnverifiedSpeculative decoding technology, in which a local small model drafts answers and a large model verifies them, can improve overall throughput speed by 2 to 4 times without degrading quality.77% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/587237API
curl https://kongchang.com/api/v1/knowledge/claims/587237MCP
get_claim(id=587237)