# VGGT-MPS

> Use this tool when you need to perform 3D vision and reconstruction tasks, such as camera pose estimation, depth prediction, and 3D point cloud generation, from image sequences. It solves problems in multi-view 3D reconstruction, providing efficient inference on Apple Silicon hardware through MPS acceleration. Ideal for researchers and developers working on 3D computer vision applications, it accepts image sequences as input and outputs 3D point clouds, camera poses, and depth maps.

Canonical page: https://skillsregistry.net/skills/jmanhype-vggt-mps  
JSON: https://api.skillsregistry.net/v1/skills/jmanhype-vggt-mps

## Description

This VGGT-MPS server provides 3D vision and reconstruction capabilities optimized for Apple Silicon through MPS acceleration, implementing the VGGT (Vision-based Geometry and Tracking Transformer) model for multi-view 3D reconstruction from image sequences. Built with PyTorch and FastMCP, it offers tools for camera pose estimation, depth prediction, 3D point cloud generation, and point tracking across video frames, featuring specialized MPS optimizations for efficient inference on Apple's Metal Performance Shaders framework. The implementation includes Gradio web interfaces, COLMAP integration for structure-from-motion workflows, and sparse attention mechanisms for scaling to large image collections, making it valuable for researchers and developers working on 3D computer vision applications who need GPU-accelerated reconstruction on Mac hardware without CUDA dependencies.

## Trust

- **Trust score (0–1):** 0.50
- **Verification tier:** unverified

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** iot-hardware
- **Updated:** 2026-04-25

## Source

- **Source listing:** [PulseMCP](https://www.pulsemcp.com/servers/jmanhype-vggt-mps)
- **Repository:** <https://github.com/jmanhype/vggt-mps>

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "jmanhype-vggt-mps"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/jmanhype-vggt-mps` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/jmanhype-vggt-mps/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt