MeshGPT.io
MeshGPT generates 3D models from text prompts or uploaded images. The publisher presents it for game props, stylized characters, product concepts, and printable meshes. You can use it to explore a shape or visual idea before checking whether the output meets the requirements of your 3D workflow.
Choose an image or a written brief
The image-to-3D workflow accepts JPG, PNG, and WebP files, including photos, sketches, and illustrations. MeshGPT advertises textured models that preserve colors and details from the input. A reference image gives the service visual guidance, but inspect surfaces that the image does not show rather than assuming hidden geometry is accurate.
Text-to-3D accepts written descriptions and promotes realistic, cartoon, and low-poly styles. The publisher also advertises PBR materials and textures. Describe the object and intended style, then check the preview for silhouette and proportion before spending time integrating a result into a scene.
The gallery includes examples labeled image-to-3D and text-to-3D. These can help you understand the types of objects the publisher chooses to demonstrate. They do not establish that every generated character is rigged, every prop fits an existing game's art direction, or every product mockup has engineering-grade dimensions.
Inspect the exported model
MeshGPT lists GLB, FBX, OBJ, and STL export formats and names Blender, Unity, Unreal Engine, and 3D printing as target workflows. Choose the format your next application can import, then verify the imported file before building a project around it. Material appearance, scale, and orientation deserve separate checks.
The site promotes clean topology and UVs, with optimized meshes and instant previews. Treat those as vendor claims and inspect geometry for your actual use. A game asset may need a particular polygon budget or deformation behavior; a print needs suitable dimensions and geometry your slicer can process. The homepage's broad use cases do not certify that a generated file meets either requirement.
For product concepts, use the model to examine form and presentation. Do not treat a visual mockup as a validated manufacturing specification or assume the service has performed structural analysis.
Confirm generation settings and rights
The FAQ says fast models typically take 20 to 30 seconds, while higher-quality generations with detailed texture and geometry settings can take one to three minutes. Those are the publisher's estimates, not guaranteed completion times.
The homepage does not give a complete price schedule, credit allocation, or detailed output-license terms. Check those conditions before a commercial project, and use input images you have permission to upload. A small test export into your real toolchain will tell you more about suitability than a speed claim alone.