Unverified50% confidenceOpinionTime unknown
The capability improvement from GPT-4 to GPT-4o is notably smaller than the leap from GPT-3 to GPT-4, indicating diminishing returns from the Scaling Law
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026
Sources
Related Claims
UnverifiedGPT-4 Turbo的输出上限只有4K token，比原版GPT-4的8K输出有所缩减81% similarUnverifiedGPT系列模型在维基百科文本上的困惑度从GPT-2的约29降至GPT-4的更低水平79% similarPartially VerifiedGPT-5.6 是对 GPT-5.5 的明显提升，但尚未达到「跨代」的程度，没有解决模型原本的弱项79% similarUnverified从GPT-3到GPT-4的进步是革命性的，但后续改进更多是在特定任务上的性能优化，而非通用能力的质的飞跃78% similarUnverifiedGPT-3到GPT-4的能力跨越发生在不到两年内78% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/59998API
curl https://kongchang.com/api/v1/knowledge/claims/59998MCP
get_claim(id=59998)