返回探索

Jev-Omni 多模态 AI:一个模型支持四种模态

Jev-Omni 以单一模型同时处理文本、图像、音频和视频,推理速度低于 100ms,弥补了其他开源模型在模态覆盖上的不足。

【New】Multimodal AI "Jev-Omni" Supports 4 Modalities with 1 Model The video version of Jev is hot. I'll try it out right away. Other OSS usually lack one or another, but this one has it all and inference under 100ms. ・Processes text/image/audio/video with 1 model ・Same level

· 0 次赞在 X 打开