Seedream 4.0

The smoke of the LLM battlefield this week has not dissipated, and Google has dropped another blockbuster. On the evening of November 20th, Beijing time, Nano Banana Pro (official version number Gemini-3-Pro-Image-Preview) was officially opened. Just three months ago, the “magic banana” that once swept the AIGC community with “everything can be done in 3D” is now making a strong return with the blessing of the powerful base of Gemini 3 Pro. Now that “Pro" is hung up…

At the end of October, whether it was a long-video B station or a short-video platform, a large number of explosive videos emerged: using the classic IP characters we know well, such as the 86th edition of "Journey to the West", they were refreshed with the blessing of AI technology, and they went into the recording studio one after another to sing in line with their respective IPS.Original song. Its mouth shape and emotional expression are highly matched with music, and with realistic video footage, it has won “three in a row with one key” time and time again. With the help of Nano Banana and Seedream 4.0, which can achieve high-fidelity picture generation, he is proficient in various music…

相信很多影像工作者近些年对 Topaz Labs 这家公司已不陌生，和一众 2024 年后围绕AI做概念就开始融资的新玩家不同，Topaz Labs 是在AI浪潮来临之前，就已经把专业技能点点满的硬核老兵——它用一系列真实可用的模型工具，在图像修复、局部放大、细节增强、去噪、插帧和视频补帧等领域打下了坚实口碑：模糊镜头能拉清晰、低分辨率素材能补成4K，尤其是…

近日，腾讯混元团队开源了其最新图像模型 Flux-1-SRPO，这款模型是基于黑森林 Flux-DEV 微调的文生图模型，其全称为 Semantic Relative Preference Optimization（语义相对偏好优化），旨在解决当前 AI 生图模型在人像生成方面的油腻皮肤质感和“AI味”常态。 针对人像生成优化，混元团队提出的创新型 Dire…

在 Google 的图片编辑模型 Nano Banana 热度尚未消退之际，字节跳动 Seed 团队于昨日发布了其全新升级的生图模型 Seedream 4.0。该模型首次支持 4K 多模态生图，灵活支持文本、图像的组合输入，实现多图融合创作、参考生图、组合生图、图像编辑等核心功能。且其推理性能较 3.0 版本提升了近 10 倍，“画”之前会先“想”，能够深度…