Unverified50% confidenceBenchmarkExact time
NCP-ArchPreview scales to 8.9 billion parameters and is trained on 5.73 trillion tokens of the Dolma-3 dataset.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/11/2026
First Seen
Valid until: 12/10/2026
Sources
Related Entities
Related Claims
UnverifiedMeta's LLaMA 3 was trained on more than 15 trillion tokens.62% similarUnverifiedResNet-18 has more than 11 million parameters, nearly 20 times the 595,000 parameters of this project.60% similarUnverifiedTabPFN在训练阶段已经见过数百万个基于贝叶斯先验分布合成生成的表格数据集60% similarUnverifiedSupernode computing power connects thousands to tens of thousands of GPUs/NPUs into a unified computing pool via high-speed interconnect networks, forming the necessary infrastructure for training large models with hundreds of billions to trillions of parameters.59% similarVerifiedMeta's Llama 3 has up to 405 billion parameters.59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/907841API
curl https://kongchang.com/api/v1/knowledge/claims/907841MCP
get_claim(id=907841)