Meshy 7 in ComfyUI: Text to 3D and Image to 3D Workflows

Meshy 7 is Meshy’s 3D generation foundation model, built around image-to-3D alignment. It reads input images at multiple scales and accepts higher-resolution input, so fine shape information survives into generation instead of being averaged out. The result is 3D geometry, color, material, and pattern that stay aligned with the source image. In ComfyUI, Meshy 7 is available through the Meshy Partner Nodes. You can generate a draft model from a text prompt or a single image, then refine it into a fully textured asset with PBR maps, rig it, and animate it, all inside one workflow. Every Meshy node exposes Meshy 7.1, Meshy 7, Meshy 6, and latest in its model option. The text-to-model template starts on Meshy 7 and the image-to-model template starts on Meshy 7.1.

​

What Meshy 7 is good at

• Text-to-3D and image-to-3D: Generate a model from a text prompt or a single input image
• Ultra Mode: An extra refinement pass for higher-fidelity geometry with finer surface detail, available on Meshy 7.1, Meshy 7, and latest models. Text to Model and Image to Model also select the pass resolution with ultra_resolution: 2k (2048³) by default, or 4k (4096³) on Meshy 7.1 and latest
• PBR texturing: Generate metallic, roughness, and normal maps with 2K to 8K base color textures
• Full asset pipeline: Refine, rig, and animate the generated model in the same workflow

​

Meshy 7 text-to-model workflow

Generate a 3D model directly from a text description. The workflow creates the draft mesh with the Meshy 7 Text to Model node, then textures, rigs, and animates it automatically.

​

Meshy 7 image-to-model workflow

Convert a single 2D image into a detailed 3D model. Meshy 7’s multi-scale image encoder preserves fine shape information from the source image, and the workflow textures, rigs, and animates the result in one pass.