Generating a 3D model in one minute sounds impressive, but the real test is in the details.

The Forge / 工坊 Models / LoRA / 模型与 LoRA
1

“One minute to generate a 3D model” is mostly clickbait—depends on what you’re using it for. For background decorations or filler props, a one-minute job gets the job done. But zoom in and it falls apart: the topology is a mess, UVs can’t be unwrapped, textures are blurry, and you can’t even tweak a single detail.

I’ve tried a few mainstream image-to-3D tools—looks impressive from the front, but spin it around and it’s a disaster, with broken faces all over the inside.

It’s great for quick concept placeholders or blocking out scenes, saves a ton of time setting up. But if you want it in a production pipeline as usable assets, you still have to retopo, repaint textures—that step you can’t skip.

2

The wiring is a total mess and the UV unwrapping won’t work—this is way too real, can’t even fix it.

3

Good enough for a wide shot, but falls apart when you zoom in, and the back view is a total mess.

4

Using rough placeholders as a good thing, saves time on building scenes.

5

You really can’t skip the retopology and re-texturing step. The promotional materials never mention that part.

6

I tried turning it to the back and it just disappeared. The front is just for show.

7

Once it goes into the official pipeline, you still gotta go back and redo everything properly. One minute is just the starting point.

8

I actually use it for background decoration, no way I’m touching serious assets with it.

9

Saved the post.