Unverified60% confidenceFactExact time
The Chinchilla optimal scaling law was proposed by DeepMind in 2022, with the core formula C≈6ND, and the condition for optimal training being D≈20N.
2
Sources
60%
Confidence
Long-term
Relevance
8/26/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedGoogle DeepMind's Chinchilla paper optimized the optimal ratio between model parameter count and training data volume.67% similarUnverified中国的DeepSeek在2025年初以极低成本训练出性能媲美顶尖模型的作品，对西方实验室的大规模训练路线构成范式挑战64% similarUnverified2022年DeepMind发布了TALM，Meta发布了Toolformer，探索让模型自主决定何时调用外部API，但依赖微调64% similarUnverifiedDeepMind于2022年推出的AlphaTensor以强化学习方法重新发现了比Strassen算法更优的矩阵乘法路径64% similarVerifiedThe Chinchilla law proposed by DeepMind further refined the optimal ratio relationship between data and parameters.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/802781API
curl https://kongchang.com/api/v1/knowledge/claims/802781MCP
get_claim(id=802781)