Rohan Paul@rohanpaul_aiFT：字节跳动据报正在预训练一个参数高达10万亿的AI模型，远超Kimi K3的2.8万亿参数规模。 FT还报道称，字节跳动一年多来一直避免蒸馏竞争对手的模型，倾向于独立开发模型。 如果上述训练成功，字节跳动将证明其能够在不依赖竞争对手模型作为教师的情况下，执行前沿规模的预训练。
FT: ByteDance is reportedly pre-training an AI model with up to 10T parameters, far exceeding Kimi K3's 2.8Trn param size.
FT also reported that ByteDance has avoided distilling rival models for more than a year, preferring independent model development.
If the reported run succeeds, ByteDance will have shown it can execute frontier-scale pretraining without leaning on a rival model as teacher.
来源：Rohan Paul · x.com
FT：字节跳动据报正在预训练一个参数高达10万亿的AI模型，远超Kimi K3的2.8万亿参数规模。 FT还报道称，字节跳动一年多来一直避免蒸馏竞争对手的模型，倾向于独立开发模型。 如果上述训练成功，字节跳动将证明其能够在不依赖竞争对手模型作为教师的情况下，执行前沿规模的预训练。
FT: ByteDance is reportedly pre-training an AI model with up to 10T parameters, far exceeding Kimi K3's 2.8Trn param size.
FT also reported that ByteDance has avoided distilling rival models for more than a year, preferring independent model development.
If the reported run succeeds, ByteDance will have shown it can execute frontier-scale pretraining without leaning on a rival model as teacher.
来源：Rohan Paul· x.com