商汤SenseNova U1.5发布8B-MoT原生统一视觉模型
SenseTime SenseNova U1.5 Brings 8B-MoT Native Unified Vision With Open Training Code
精选理由
商汤发布了SenseNova U1.5，这是一个8B-MoT原生统一视觉模型，支持4K生成，还开放了训练代码。
商汤的SenseNova U1.5是一款8B-MoT原生统一视觉模型，无需外部编码器/VAE，支持4K生成，并开放SFT/RL/蒸馏训练代码。
原文 · pandaily
SenseTime SenseNova U1.5 Brings 8B-MoT Native Unified Vision With Open Training Code
SenseTime's SenseNova U1.5 is an 8B Mixture-of-Transformers native unified vision model—no external encoder/VAE—with up to 4K generation and open SFT/RL/distillation training code.
