Hy3：3D人体姿态生成新SOTA
推荐指数 57.0 NO. 016 · 2026.07.10
发布2026/07/09Score216Comments60
为什么值得看
Hy3是新的3D人体姿态生成模型，支持从文本或单张图片生成高质量3D人体动作。对做虚拟人、游戏动画、运动分析的AI工程师来说，这是目前开源方案里效果最接近商用的选择之一。
编辑判断
做3D虚拟人的团队之前大多依赖SMPL系模型加手动关键帧，或者牺牲质量用轻量方案。Hy3的亮点是统一了文本驱动和单图驱动两条路线，生成结果的物理合理性比之前的MDM、MotionDiffuse有明显提升，关节穿透和脚滑问题少很多。
游戏和短剧赛道最近对低成本动捕替代方案需求很旺，这个工具链成熟度够的话，可能直接吃掉一部分惯性动捕设备的市场。建议优先测它的单图驱动分支，比文本可控性更强，落地更快。
社区反馈
意见分歧 60 条评论
核心争论：Hy3是否被过度炒作，实际性能能否匹敌DeepSeek V4系列及GPT-5.4
相关内容
I tried out the model it's pretty great, better than ~~gpt5.4~~ gpt-5.4-mini perhaps, atleast close enough to sonnet 5 in performance that I didn't notice much of a gap. Not really at gpt 5.5 tier though, and probably below glm 5.2... But most of all it just works for me for most things I tried and
I think you’ve got the models wrong…gpt-5.4? I doubt there is any open source mode matching it. Maybe in a year
Yeah I meant gpt-5.4-mini, but GLM 5.2 is pretty close to gpt-5.4 base, and much better than it when it comes to design stuff.