Unverified90% confidenceBenchmarkExact time
The 70-billion-parameter Chinchilla model, trained on 1.4 trillion tokens, outperformed Gopher, which has 280 billion parameters but was trained on only 300 billion tokens.
1
Sources
90%
Confidence
Long-term
Relevance
8/19/2026
First Seen
Sources
Rethinking Scaling Laws: Parameters Are Not the Only Answer
twitterjietang8/19/2026
Related Entities
Related Claims
UnverifiedChinchilla surpassed Gopher, which has 280 billion parameters, on multiple benchmarks with only 70 billion parameters.71% similarUnverifiedChinchilla模型以70B参数配合4倍数据量即可超越280B参数的Gopher模型性能67% similarUnverifiedAccording to the Chinchilla scaling law, the optimal training data volume for a 30B model is approximately 600B–1T tokens.65% similarPartially VerifiedTongyi Qianwen released the Qwen 3.8 Max model, with a parameter scale as high as 2.4 trillion.65% similarUnverifiedQwen Coder adopts a Mixture of Experts (MoE) architecture with 480 billion total parameters, activating only 35 billion parameters during inference.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/796926API
curl https://kongchang.com/api/v1/knowledge/claims/796926MCP
get_claim(id=796926)