How an agent built a 3D Paris gallery by chaining two HuggingFace Spaces
Original title:How an Agent Built a 3D Paris Gallery by Chaining Two Hugging Face Spaces
AI summary
A coding agent called two Spaces on HuggingFace to build from scratch an interactive gallery displaying 3D Gaussian splats of Paris landmarks. The agent first used ideogram-ai/ideogram4 to generate black-background images of each monument, then reconstructed 3D Gaussian splats (.ply) from single images via VAST-AI/TripoSplat, automatically completed coordinate system correction, framing, and compression to .ksplat (about 3x smaller), and built a Three.js-based scroll-switching, drag-to-rotate viewer, finally deploying it as a static Space. The entire process required no client-side libraries; each Space exposed callable APIs via `agents.md`.
Key Takeaways
- A coding agent called two Spaces on HuggingFace to build from scratch an interactive gallery…
- The agent first used ideogram-ai/ideogram4 to generate black-background images of each…
- ply) from single images via VAST-AI/TripoSplat,automatically completed coordinate system…
- ksplat (about 3x smaller),and built a Three.
- js-based scroll-switching,drag-to-rotate viewer,finally deploying it as a static Space.
Quick Facts
- Published
- Updated
- Source
- Hugging Face 社区博客(混合发现)
- Category
- Tutorials & Tips
- AI score
- 76
Hugging Face 把 agents.md 做成每个 Space 的标准说明书,agent 能直接读懂并链式调用图像和 3D 模型,这篇用 3D 巴黎画廊 demo 告诉你这事儿已经跑通了,做 AI 工具链的可以立刻照着试。
Related entities
Source & copyright
This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.
How to Cite
When citing this article, keep the title, date, and permalink, and credit Orange News as the source.
Orange News: How an agent built a 3D Paris gallery by chaining two HuggingFace Spaces. June 9, 2026. https://new.nat6.net/en/news/cmq6l2wms09elsl5ipt8ah0ij/