
The ease with which you can generate everything you imagine. What I like most about Meshy is that it allows you to turn an idea or a reference image into a usable 3D model in a very short time. I mainly use it to create characters and elements for a video game, and it has allowed me to go from concept to prototype without needing weeks of manual modeling.
The interface is clear and the process is well organized, even for someone who is not a 3D modeling specialist. It is easy to generate different versions, preview them, apply textures, and download the result. Exporting to standard formats like GLB greatly facilitates integration with engines like Godot, although some models need post-adjustments of materials, scale, rigging, or animations.
Overall performance is good and generations are usually completed quickly. The AI interprets visual references and prompts quite well, especially when using clear images and characters in neutral poses. I also value the ability to generate textures and animations within the same platform. You don't always get the perfect result on the first try, but being able to create variations quickly greatly accelerates the creative process.
As for the price, I consider it offers a good return on investment for independent developers and small studios. It significantly reduces the time and cost needed to produce 3D prototypes. The credit system requires careful planning of generations, but it is still more economical than modeling each resource from scratch or outsourcing all models externally.
The onboarding process is accessible and the documentation and tutorials allow you to get started quickly. Overall, Meshy does not completely replace the work of a 3D artist, but it works very well as a production and prototyping tool. In my case, it has made it feasible to develop and test characters for a video game much faster, maintaining a flexible workflow between AI generation, retouching, and engine integration. Review collected by and hosted on G2.com.
What I like least about Meshy is the lack of consistency between generations. Even using the same reference image and a detailed prompt, the AI can modify important features, proportions, clothing, or small elements of the character. This forces you to spend several attempts and credits to achieve an adequate result.
Although the interface is simple, it would be useful to have more precise controls to preserve visual identity, correct only one area of the model, or lock already approved parts. Rigging and animations may also require post-adjustments: some deformations do not appear natural, and the names or organization of exported clips do not always facilitate their direct integration into engines like Godot.
Overall performance is good, but certain processes can take time, and the final result does not always justify the credits consumed. Therefore, the price can be high when many generations need to be repeated to correct small errors. A more flexible credit system for failed attempts would significantly improve the return on investment.
Exporting in standard formats works well, although models still often need additional work on materials, scale, geometry, rigging, or animations before they can be used in production. I would also appreciate more specific documentation on these issues and support more oriented towards workflows with game engines.
Meshy is a very useful tool for creating prototypes, but it still needs greater control, consistency, and precision for the results to go directly into production without so many manual corrections. Review collected by and hosted on G2.com.