[Integrations](https://spiral.ai/integrations) › [Codex](https://spiral.ai/integrations/codex) › [Meshy 7.1 Multi-Image to 3D](https://spiral.ai/integrations/codex/meshy-7-1-multi-image-to-3d)

# How to Use Meshy 7.1 Multi-Image to 3D in Codex

Meshy 7.1 Multi-Image to 3D turns reference photos into detailed meshes for product visualization, props, and sculpting. It uses additional angles to capture back and side details that a single image leaves unseen.

|  |  |
| --- | --- |
| Model | Meshy 7.1 Multi-Image to 3D |
| Available through | fal |
| Category | 3D |
| Price | `should_texture`: false $1.04, true $1.56 per model. `enable_rigging` adds $0.26 per model. `enable_rigging` and `enable_animation` add $0.156 per model. |
| Returns | 3d, image |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with Meshy 7.1 Multi-Image to 3D in Codex

- Use Meshy 7.1 Multi-Image to 3D with the front, side, and back product images in this project, front first. Generate a detailed mesh and check its silhouette against all three.
- Use Meshy 7.1 Multi-Image to 3D to reconstruct the prop shown in the project's reference images. Keep only views with a matching pose and lighting.
- Split the project's character turnaround into separate angle images, then use Meshy 7.1 Multi-Image to 3D with the front view first. Flag thin hair details for inspection.
- Use Meshy 7.1 Multi-Image to 3D on the project's object photos and remesh toward 20,000 faces. Report the actual face count and inspect the silhouette.
- Use Meshy 7.1 Multi-Image to 3D with the project's lamp references. Request PBR maps and use this texture instruction: brushed brass body, matte black shade.

## How to set up Meshy 7.1 Multi-Image to 3D in Codex

Pick the way that suits you. Each takes about two minutes.

### Use Spiral's MCP server

1. **Add the hosted server**

```bash
codex mcp add spiral --url https://api.spiral.ai/mcp
```

2. **Sign in**: Codex opens the browser: sign in to Spiral and click Authorize. No API key. `codex mcp list` then shows Spiral.

```bash
codex mcp login spiral
```

### Use the Spiral skill

1. **Send Codex this message**: Your agent reads Spiral's skill and sets itself up.

```
set up https://spiral.ai/SKILL.md
```

2. **Sign in or paste a key**: Approve Spiral in the browser, or give the agent a key from the console.

### Use the Spiral CLI

1. **Install and sign in**: Installs the Spiral CLI, signs you in and adds its skill to your coding agents.

```bash
npx -y spiral-cli@latest init
```

### Call the Spiral API

1. **Create an API key**: In the console, under API keys.

2. **Run Meshy 7.1 Multi-Image to 3D**: One request; the result comes back by polling or a webhook.

```bash
curl https://api.spiral.ai/v1/runs \
  -H "Authorization: Bearer $SPIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "provider": "fal",
    "model": "meshy/v7.1/multi-image-to-3d",
    "input": {
      "image_urls": [
        "<image_urls>"
      ]
    }
  }'
```

Then paste one of the prompts above: Codex finds Meshy 7.1 Multi-Image to 3D on Spiral, shows the price and runs it.

Setup guide in the docs: https://spiral.ai/docs/quickstarts/codex

## Meshy 7.1 Multi-Image to 3D tools and prices

- [Meshy 7.1 Multi-Image to 3D](https://spiral.ai/tools/fal/meshy/v7.1/multi-image-to-3d) (fal): `should_texture`: false $1.04, true $1.56 per model. `enable_rigging` adds $0.26 per model. `enable_rigging` and `enable_animation` add $0.156 per model.

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| topology | quad, triangle | triangle | Specify the topology of the generated model. Quad for smooth surfaces, Triangle for detailed geometry. |
| target_polycount | 100 to 300000 | 30000 | Target number of polygons in the generated model |
| symmetry_mode | off, auto, on | auto | Controls symmetry behavior during model generation. |
| should_remesh |  | true | Whether to enable the remesh phase. When false, returns triangular mesh ignoring topology and target_polycount. |
| should_texture |  | true | Whether to generate textures. |
| texture_model | meshy-7, meshy-n1 | — | Model that textures the generated mesh from several views, in the same task. Leave unset to use the endpoint model's default texture model. Ignored when should_texture is false. meshy-n1 makes a 2K base color texture from reference images only: it rejects texture_prompt, enable_pbr, and quad topology with remeshing. With meshy-n1, Meshy textures from texture_image_urls or texture_image_url, or from the input images when neither is set. |

## What can you do with Meshy 7.1 Multi-Image to 3D in Codex?

Turn project reference images into detailed meshes for product and prop reviews. Your coding agent can organize the views and compare the generated shape against the source images.

### Product shape reviews

Use matching front, side, and back photos to create a mesh for reviewing a product from several angles.

### Detailed prop studies

Turn reference images of a prop into a detailed mesh, then compare its silhouette against each supplied view.

### Character turnaround reconstruction

Split character artwork into separate angle images and use them to guide the front, sides, and back of one model.

### Sculpting starting meshes

Generate a detailed base from visual references, then inspect thin features before further sculpting.

### Material studies

Create a textured object from photos and request PBR maps to study metallic and rough surfaces in a render.

### Reduced-detail review assets

Remesh a reference-based object toward a chosen face count, then check which silhouette details survived.

## Overview

- [fal](https://spiral.ai/providers/fal)
- [All Codex guides](https://spiral.ai/integrations/codex)

## Meshy 7.1 Multi-Image to 3D in other agents

- [Amp](https://spiral.ai/integrations/amp/meshy-7-1-multi-image-to-3d)
- [ChatGPT](https://spiral.ai/integrations/chatgpt/meshy-7-1-multi-image-to-3d)
- [Claude](https://spiral.ai/integrations/claude-desktop/meshy-7-1-multi-image-to-3d)
- [Claude Code](https://spiral.ai/integrations/claude-code/meshy-7-1-multi-image-to-3d)
- [Cline](https://spiral.ai/integrations/cline/meshy-7-1-multi-image-to-3d)
- [Cursor](https://spiral.ai/integrations/cursor/meshy-7-1-multi-image-to-3d)
- [Devin](https://spiral.ai/integrations/devin/meshy-7-1-multi-image-to-3d)
- [dots](https://spiral.ai/integrations/dots/meshy-7-1-multi-image-to-3d)
- [Factory Droid](https://spiral.ai/integrations/factory-droid/meshy-7-1-multi-image-to-3d)
- [Gemini](https://spiral.ai/integrations/gemini/meshy-7-1-multi-image-to-3d)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/meshy-7-1-multi-image-to-3d)
- [Google Antigravity](https://spiral.ai/integrations/antigravity/meshy-7-1-multi-image-to-3d)
- [goose](https://spiral.ai/integrations/goose/meshy-7-1-multi-image-to-3d)
- [Grok](https://spiral.ai/integrations/grok/meshy-7-1-multi-image-to-3d)
- [Grok Build](https://spiral.ai/integrations/grok-build/meshy-7-1-multi-image-to-3d)
- [Hermes](https://spiral.ai/integrations/hermes/meshy-7-1-multi-image-to-3d)
- [Instinct](https://spiral.ai/integrations/instinct/meshy-7-1-multi-image-to-3d)
- [Manus](https://spiral.ai/integrations/manus/meshy-7-1-multi-image-to-3d)
- [Muse](https://spiral.ai/integrations/muse/meshy-7-1-multi-image-to-3d)
- [Muse Code](https://spiral.ai/integrations/muse-code/meshy-7-1-multi-image-to-3d)
- [n8n](https://spiral.ai/integrations/n8n/meshy-7-1-multi-image-to-3d)
- [Notion](https://spiral.ai/integrations/notion/meshy-7-1-multi-image-to-3d)
- [OpenClaw](https://spiral.ai/integrations/openclaw/meshy-7-1-multi-image-to-3d)
- [opencode](https://spiral.ai/integrations/opencode/meshy-7-1-multi-image-to-3d)
- [Perplexity](https://spiral.ai/integrations/perplexity/meshy-7-1-multi-image-to-3d)
- [Raycast](https://spiral.ai/integrations/raycast/meshy-7-1-multi-image-to-3d)
- [VS Code](https://spiral.ai/integrations/vscode/meshy-7-1-multi-image-to-3d)
- [Windsurf](https://spiral.ai/integrations/windsurf/meshy-7-1-multi-image-to-3d)
- [Zapier](https://spiral.ai/integrations/zapier/meshy-7-1-multi-image-to-3d)

## More Codex guides

- [Meshy 7.1 Image to 3D](https://spiral.ai/integrations/codex/meshy-7-1-image-to-3d)
- [Meshy 7.1 Text to 3D](https://spiral.ai/integrations/codex/meshy-7-1-text-to-3d)
- [Meshy 6 Preview Text to 3D](https://spiral.ai/integrations/codex/meshy-6-preview-text-to-3d)
- [Anthropic Claude](https://spiral.ai/integrations/codex/anthropic)
- [Meshy 6 Preview Image to 3D](https://spiral.ai/integrations/codex/meshy-6-preview-image-to-3d)
- [AssemblyAI](https://spiral.ai/integrations/codex/assemblyai)

## Questions about Meshy 7.1 Multi-Image to 3D in Codex

**Can Codex use Meshy 7.1 Multi-Image to 3D?**

Yes. Add Spiral to Codex as an MCP server, and Codex can find and run Meshy 7.1 Multi-Image to 3D like any built-in tool.

**How much does Meshy 7.1 Multi-Image to 3D cost in Codex?**

`should_texture`: false $1.04, true $1.56 per model. `enable_rigging` adds $0.26 per model. `enable_rigging` and `enable_animation` add $0.156 per model. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Codex plan is separate.

**Do I need a fal account to use Meshy 7.1 Multi-Image to 3D in Codex?**

No. Spiral runs Meshy 7.1 Multi-Image to 3D on its own fal account. You sign in to Spiral once and never paste a key.

**How long does it take to set up Meshy 7.1 Multi-Image to 3D in Codex?**

About two minutes: add the hosted server, then sign in.

**Does Codex wait while Meshy 7.1 Multi-Image to 3D runs?**

Meshy 7.1 Multi-Image to 3D runs in the background. Codex starts the run, checks on it and gives you the 3d, image when it's ready.

**What is Meshy 7.1 best at?**

Detailed meshes that closely follow visual references. Multiple views make it useful when a product or prop needs recognizable sides and a back.

**Do more reference images always improve the result?**

No. One clear image can beat several conflicting images, so keep the object, pose, and lighting consistent across views.

**Which details need the closest inspection?**

Thin features such as hair strands and jewelry chains are difficult to reconstruct from one image. Inspect those areas before accepting the mesh.

**Does the polygon target guarantee an exact output count?**

No. Remeshing aims for your target, but the actual face count depends on the geometry.

**What do PBR maps add?**

They add metallic, roughness, and normal maps to the base color texture. Use them when surface materials matter in your render.

