jietang· @jietang · X·· 2026-09-10AI 评分29
AI 导读
唐杰指出确定模型最优规模并不简单，需权衡数据量、激活参数量、环境数量和目标推理成本，性能还受多种因素影响。其引用讨论称 Fable 约为 2-2.5T 参数而非 10T，Kimi K3 为 2.8T 参数、训练约用 20-30k Blackwell 等效算力，能力接近 Fable 5（非 5.1）。
正文 · 原文
Are you sure? Finding the optimal model size is tricky: data volume, active parameter count, the number of environments, and the target inference cost. Model performance also depends on many other factors, each introducing its own variability.
Fable is probably ~2-2.5T parameters, not 10T. Kimi K3 is 2.8T params, trained on maybe 20–30k Blackwell-equivalents. It lands within spitting distance of Fable 5 in terms of capabilities (5, not 5.1). Anthropic has far more compute than Moonshot, better rl environments, better architecture and better optimizers and all of that adds to capability per parameter. So if Fable is only slightly ahead of K3 with this in mind, it's almost certainly a smaller model. GPT-5.5 and 5.6 are smaller still (I'll say more on that later)在 X 查看被引用的帖子
来源：jietang · x.com