Create

Sign in to ReadmeX

or

How we handle your data: Privacy Policy

ReadmeX

Your AI reading companion

What would you like to catch up on?

Pick a briefing, or ask your own question.

Your community and people briefings stay personal to you.

Story

Hi3D V1.0 open-sources a production-tested image-to-3D model

AI summary

Shumei Wanwu has open-sourced the code and weights for Hi3D V1.0, an image-to-3D geometry refinement model built on Sparc3D and tested in production workflows before release. According to the vendor's technical report, the model was trained on about two million selected 3D assets and supports effective surface extraction resolutions up to 1536³. The same report lists a 0.75 multi-view fidelity score, 0.841 geometric detail richness and a 90% slicing success rate, though these figures have not been independently verified in the supplied reports.

In-depth insight

The release gives developers and enterprises an openly available foundation for building and deploying high-resolution image-to-3D workflows, while Hi3D Agent provides a separate creator-facing layer for modifying and preparing generated assets.

原创 关注AI三维的 2026-10-26 14:13 北京

AI 3D,从开放模型,生长出无限可能

编辑|冷猫

自从电影《牛来》现象级爆火,那只造型粗糙、通体发黄的小牛,就成了不少人提起三维建模时的第一反应。

#Hi3D V1.0#Shumei Wanwu#Sparc3D#Hi3D Agent#Hi3D V3.0#GitHub#Hugging Face

Original · 13 min readRead the original
Group chat

No comments yet. Start the conversation.