Meshy 7 · Image to 3D

ABOUT THE WORKFLOW

Turn an Image Into a 3D Model Upload a photo, product shot, or concept image of any object. Meshy 7 reconstructs it as a textured 3D mesh you can rotate in the built-in preview, then export in the format your pipeline needs. One image in, one 3D model out.

Partner node. This workflow calls an external API, so each run uses credits from your API wallet. No API key needed. Floyo handles the connection.

Model

• Meshy 7 by Meshy. The latest image-to-3D foundation model, focused on alignment between your reference image and the generated geometry.

HOW IT WORKS

Step 1. Upload your image A photo, product shot, render, or concept illustration of the object you want as a 3D model. One clear object on a clean background works best. Works great with: product photos · concept art · character renders · props

Step 2. Hit run Meshy 7 reads your image and generates a 3D mesh with textures that match the reference. Generation takes about a minute.

Step 3. Preview your model The finished model loads in a 3D viewer right in the workflow. Rotate it, inspect the geometry, and decide whether to export or re-run with a different image.

Step 4. Download and export Grab the model file and bring it into your 3D pipeline. Ready for: Blender · Unity · Unreal Engine · 3D printing slicers

First time? Leave every setting as-is. The defaults (standard · triangle · 30K polys · textures on) are the right starting point for almost everyone.

RECOMMENDED SETTINGS

Quick-start guide. Find the goal that matches yours and copy the settings.

• Standard 3D from a photo (most people) — Standard model type · triangle · 30,000 polys · textures on. The right starting point for almost everyone.

• Game-ready topology — Switch model type to "smart topology." The mesh comes back with cleaner edge flow suited for a game engine, without a manual retopology pass.

• Highest quality for a hero asset — Turn ultra mode on. Sharpest textures and tightest geometry, at 5 extra credits per run.

• Need quads for editing in Blender or Maya — Switch topology from triangle to quad. Quads subdivide and deform more predictably in DCC tools.

• Want untextured geometry only — Turn "should texture" off. You get the raw mesh without generated surface textures, useful when you plan to paint or bake your own.

• Override the surface look — Add a texture prompt to guide the texturing stage away from the source image. Write something like "hand-painted cartoon style" or "brushed steel with scratches."

• Reducing polycount for mobile or web — Lower target polycount toward 5,000 to 10,000. Range goes from 100 to 300,000.

Input image: One clear object on a clean or neutral background, with even lighting. Cluttered backgrounds, multiple overlapping objects, and extreme angles make reconstruction harder. Product photos and concept renders on white backgrounds produce the best results.

LEARN

📹 Videos

• Intro to Floyo

• ComfyUI 101 Free Course ft. Sebastian Kamph

• Floyo 101 for Team Collaboration

✨ Quick links

• Help docs

• LoRA Training guide

• Discover more workflows

USE CASES

🛍️ E-commerce and Product Visualization Turn a product photo into a 3D model customers can rotate. Export as GLB for a web 3D viewer or USDZ for AR try-on on Apple devices.

🎮 Game Developers Convert concept art or reference photos into placeable 3D assets. Use smart topology mode for cleaner meshes that are closer to game-ready out of the box.

🎨 Concept Artists and 3D Artists Skip the blockout stage. Upload a 2D concept and get a 3D starting point to refine in Blender, Maya, or ZBrush.

🖨️ 3D Printing Photo of a figurine, prop, or part? Turn it into a printable mesh. Export as STL or 3MF and send it to your slicer.

🥽 AR/VR and Spatial Computing Create 3D assets for augmented reality scenes, virtual showrooms, or spatial apps. Export as USDZ for Apple platforms or GLB for web-based viewers.

WHAT WORKS BEST / WHAT TO AVOID

✅ Works great

• Single object on a clean background

• Product photos with even lighting

• Concept art and character renders

• Front or three-quarter angle views

⚠️ May produce softer results

• Cluttered backgrounds with multiple objects

• Extreme angles or heavy foreshortening

• Flat illustrations with no depth cues

• Transparent or reflective objects (glass, mirrors)

FAQ

What is Meshy 7 image to 3D? Meshy 7 is the latest 3D foundation model from Meshy, released August 2026. Its image-to-3D mode takes a single reference photo and generates a textured 3D mesh that aligns closely with the original. It focuses on geometry alignment and surface detail, scoring higher on fidelity benchmarks than previous Meshy versions.

How is image-to-3D different from text-to-3D? Image-to-3D starts from a photo or concept image you upload. The model reads the shape, lighting, and surface details from your image and reconstructs them as a 3D mesh. Text-to-3D starts from a written description and generates the object from scratch. Use image-to-3D when you already have a visual reference. Use text-to-3D when you only have an idea.

What kind of images work best with Meshy 7? A single object on a clean or neutral background with even lighting. Product photos, concept renders, and character illustrations work well. Avoid cluttered scenes, multiple overlapping objects, extreme angles, and heavily transparent or reflective surfaces. The cleaner your input, the closer the 3D output matches.

What file formats can I export from Meshy 7? The workflow generates a GLB file by default. From Meshy you can export as GLB, FBX, OBJ, USDZ, STL, BLEND, 3MF, or DXF. GLB works for game engines and web. FBX suits animation pipelines. STL and 3MF are for 3D printing. USDZ is for Apple AR.

Can I use Meshy 7 outputs commercially? Meshy is a closed-source, proprietary API with credit-based pricing. Commercial use is allowed on paid Meshy plans. Free-tier outputs carry a CC BY 4.0 license requiring attribution. Review Meshy's current terms of service for your plan tier before shipping assets in a commercial product.

Can I control the mesh topology and polycount? Yes. Target polycount ranges from 100 to 300,000, defaulting to 30,000. You can choose triangle or quad topology, and switch between standard, low-poly, or smart-topology mesh types. Smart topology produces cleaner edge flow for game-ready assets without a manual retopology pass.

How to run Meshy 7 image to 3D online? You can run Meshy 7 image to 3D online through Floyo. No installation, no setup, no API key to wire up. Open the workflow in your browser, upload your image, and hit run. Free to try.

WHY FLOYO?

Floyo is the only platform with team collaboration for ComfyUI in the browser. You run workflows with no install. You share run history, assets, and models across your team. You pay only when you generate. Floyo supports open-source and closed-source models.

A designer runs a generation and likes the result. A teammate opens that exact run from shared history and keeps going. No file handoffs. No version confusion.

For studios and enterprise teams, Floyo adds private workspaces, pooled resources, and a team usage dashboard. Other ComfyUI cloud tools run for one person at a time. Floyo runs for the whole team, with transparent per-generation costs.

Ready to try it? Upload a photo of any object and hit run. The settings are already set.

→ Launch Workflow, Free

Questions? Watch the free course or check the FAQ above.