Unverified50% confidenceBenchmarkExact time
模拟存内计算的功耗通常比同等算力的数字芯片低5-10倍
1
Sources
50%
Confidence
Long-term
Relevance
8/25/2026
First Seen
Sources
Related Entities
Related Claims
Unverified当权重仅为-1、0、+1时，矩阵乘法可完全用加减法代替，理论上可将神经网络推理功耗降低5-10倍70% similarUnverified轻量版模型单次推理的算力消耗往往只有旗舰版本的十分之一甚至更低70% similarUnverified效率型模型以远低于旗舰模型的推理成本（通常低5-10倍）提供接近旗舰模型80-90%的能力表现69% similarUnverifiedDuring the Decode phase of large models, because each step depends on the result of the previous step and can only be executed serially, hardware utilization is typically below 10%.66% similarUnverifiedDuring the autoregressive decoding phase, GPU compute utilization is often less than 10%, making it a memory-intensive operation.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/798238API
curl https://kongchang.com/api/v1/knowledge/claims/798238MCP
get_claim(id=798238)