基于OmniLottie创新点换强基座模型，并且迁移Lottie词表扩展规则。冻结主干参数，在attention/MLP指定投影层插入LoRA，同时训练新增Lottie token对应的embedding 和 LM_head 行；训练目标是条件输入到Lottie token的自回归生成，只对Lottie target段算loss

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support