Unverified50% confidenceFactExact time
For a configuration of 1.6 trillion total parameters and 49 billion activated parameters, only about 3% of parameters participate in computation during actual inference.
1
Sources
50%
Confidence
Long-term
Relevance
9/12/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedGrok 1 的3140亿总参数中每次推理仅激活约25%，实际推理算力接近650亿参数规模的密集模型78% similarUnverifiedA model with 100 billion parameters requires approximately 80 GB of VRAM to process a single inference request, while a model with tens of billions of parameters requires only about 20 GB.77% similarUnverifiedTongyi Qianwen 3.8 has approximately 2 trillion parameters, and Kimi has approximately 2 trillion parameters.77% similarUnverifiedKimi 2.5 computes only approximately 30 billion parameters per forward pass, despite having a total parameter count of 1 trillion.71% similarVerifiedIf a MoE model activates only approximately 1/8 of its parameters per inference, the actual computational overhead may be equivalent to that of a dense model with tens of billions of parameters.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/912187API
curl https://kongchang.com/api/v1/knowledge/claims/912187MCP
get_claim(id=912187)