Meshy
Overview
Meshy is an AI tool that generates 3D models from text or images. Targeted at game developers, 3D artists, and product designers, it outputs textured, rig-ready 3D assets in seconds to a few minutes. In addition to the browser-based web app, Meshy provides official plugins for Blender, Unity, Unreal, and Maya so it can slot directly into existing 3D pipelines.
The current foundation model is Meshy 7, which went live on August 10, 2026. It is an image-to-3D model built around a single standard — the 3D result should agree with the image the user brought — and Meshy published the geometry-alignment benchmark behind that claim, scoring geometry directly against reference models held out of training rather than by human preference. In single-view tests it posted 81.0% on overall proportion, 79.7% on spatial distribution, and 59.8% on surface details, leading all four competing 3D foundation models on every metric.
Key Features
1. Text to 3D / Image to 3D
Both Text to 3D (generate a model from a text prompt) and Image to 3D (reconstruct a mesh from a single image) are supported. Generation usually follows a two-stage flow — preview mesh first, then refine — letting you upscale only when the draft looks promising. Meshy 7 is open to every subscription tier, but Ultra Mode currently supports single-view generation only, with multi-view input arriving later. Outside Ultra Mode, Image to 3D also offers Multi-View Image to 3D: feed up to four photos of the same object from different angles and Meshy rebuilds both geometry and base-color texture (up to 8K) from every view, sharpening accuracy on backs and sides a single photo can't show. The same capability is exposed via the API through the multiview_image_urls parameter.
2. PBR Textures and Retopology
Meshy auto-generates PBR textures (albedo, normal, roughness, metallic) and can retopologize the mesh into quad polygons that are ready for animation or sculpting. A Low Poly Mode outputs lightweight game-ready meshes directly.
3. Text to Texture
A dedicated Text to Texture mode applies text-driven textures to an existing mesh — convenient for restyling legacy assets without touching the geometry.
4. Rigging and Animation
Meshy can auto-rig humanoid characters and preview generic motion clips, compressing the early stages of game and avatar production.
5. 3D Agent (conversational modeling)
Meshy 3D Agent, released in July 2026, turns a conversation, a photo, a sketch, or a line of text into a finished model. Rather than generating immediately it brainstorms several creative directions first, lets you refine the favorite through chat, and then converts it into a downloadable FBX / OBJ / GLB / STL asset — no prior 3D experience required.
6. 3D Printing Output
Meshy natively supports the multi-color 3MF format for 3D printing, reporting a 97% slicer pass rate on complex figurines and cutting pre-print preparation from about 30 minutes to under one minute.
7. DCC Plugins
Official plugins for Blender, Unity, Unreal, and Maya let you call the Meshy API directly from inside your DCC tool of choice.
Plans & Pricing
The lineup runs Free / Pro / Premium / Ultra / Studio / Enterprise. The free tier grants 100 credits per month, its output is licensed CC BY 4.0 (commercial use allowed with attribution), and downloading models generated with Meshy 7 requires Pro or above.
| Plan | Price/mo | Highlights |
|---|---|---|
| Free | $0 | 100 credits/mo, no credit card required |
| Pro | $20 ($240/yr) | 1,000 credits/mo, 60% faster generation, API, 10 concurrent tasks |
| Premium | $40 | 3,000 credits/mo |
| Ultra | $100 | 8,000 credits/mo, top individual tier for heavy generation |
| Studio | $70 | 1 member included ($10/mo per extra member), 5,500 shared credits/mo |
| Enterprise | Contact | Custom terms and support |
Pricing can change with FX rates, promotions, or plan revisions — check the official site for the latest.
Typical Use Cases
- Game dev: mass-produce prototype props, environment assets, and monsters
- Metaverse / VR: first-pass avatars and scene objects
- Concept art: turn 2D images into rough 3D mockups for composition studies
- Product design: rapid prototyping of figures, furniture, and gadgets
- 3D printing: figures and goods produced through multi-color 3MF output
- Education: teaching materials for students with no 3D modeling background
Strengths
- ✅ Accepts both text and image input
- ✅ Publishes a measurable image-to-3D alignment score with Meshy 7
- ✅ PBR textures + quad retopology — ready for games and video
- ✅ Rich DCC plugin ecosystem (Blender / Unity / Unreal / Maya)
- ✅ Two-stage preview → refine flow saves credits
- ✅ API available for integration into custom pipelines
Weaknesses
- ❌ Struggles with complex mechanical parts and thin-sheet geometry; meshes can break
- ❌ High-resolution refinement consumes a lot of credits
- ❌ Weak on hair, transparent objects, and cloth simulation
- ❌ Downloading Meshy 7 output requires a Pro plan or above, and free-tier output is CC BY 4.0 so commercial use needs attribution
Official Resources
- Website: https://meshy.ai
- Docs: https://docs.meshy.ai
- Pricing: https://meshy.ai/pricing