Unverified50% confidenceFactExact time
For each token generated or processed, one forward pass must be executed, and its computational cost is roughly proportional to the number of model parameters.
1
Sources
50%
Confidence
Long-term
Relevance
10/16/2026
First Seen
Sources
Related Entities
Related Claims
Unverified大模型推理的自回归生成需要逐Token串行输出，生成N个Token需要N次前向传播78% similarUnverifiedEach time a token is generated via a forward pass, the model must go through a complete multi-head attention computation and feed-forward network layers, with the amount of computation determined by the model's parameter scale78% similarUnverified输出生成是严格串行的自回归过程，每生成一个token都需要一次完整的模型前向传播78% similarUnverifiedAn autoregressive model requires only one forward pass to generate each token.75% similarUnverifiedEach model invocation calculates both input tokens and output tokens simultaneously, and both together determine the cost of a single call.74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1049577API
curl https://kongchang.com/api/v1/knowledge/claims/1049577MCP
get_claim(id=1049577)