Unverified50% confidenceFactExact time
The prefill phase of large model inference is compute-intensive (matrix-matrix multiplication) and bound by compute; the decode phase is memory-intensive (matrix-vector multiplication) and bound by memory bandwidth.
1
Sources
50%
Confidence
Long-term
Relevance
10/18/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedLLM 推理分为预填充阶段（Prefill，计算密集型）和解码阶段（Decode，内存带宽密集型）两个阶段77% similarUnverifiedThe prefill phase of large language model inference is computationally intensive and can fully utilize a large number of parallel cores.77% similarUnverifiedThe Prefill stage of LLM inference is a compute-intensive operation whose primary bottleneck is GPU/NPU compute (FLOPS).76% similarVerifiedThe inference stage of large models is memory-bound (limited by memory bandwidth); generating each token requires reading the entire model weights from VRAM, resulting in extremely low computational intensity.74% similarUnverifiedThe Prefill stage is the first step of large language model inference, in which the model performs a complete forward computation in parallel over all input tokens to generate the KV Cache for every layer.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1062225API
curl https://kongchang.com/api/v1/knowledge/claims/1062225MCP
get_claim(id=1062225)