---
title: "Spark Environment Setup - Air Market"
description: "Setting up ML on this particular NVIDIA hardware tends to stall on install errors that don't look like anything you've hit before. This walks through getting a working training and inference environm…"
canonical: "https://airmarket.ai/app/spark-environment-setup"
---

[Air Market](/) / [Data & Analytics](/category/data) / Spark Environment Setup

Skill by [Seth Hobson](/@wshobson)

# Spark Environment Setup

Get PyTorch running on DGX Spark instead of fighting wheels

## What this does

Setting up ML on this particular NVIDIA hardware tends to stall on install errors that don't look like anything you've hit before. This walks through getting a working training and inference environment on it, including the errors specific to its architecture. Reach for it when the install won't go through, or when you're deciding between a container and installing packages directly.

### Top 3 things it does for you

- Install PyTorch, Unsloth, TRL, or vLLM on DGX Spark
- Sort out libcudart and wheel-ABI errors on aarch64
- Decide between an NGC container and plain pip installs

## Information

**Spark Environment Setup at a glance**

|  |  |
| --- | --- |
| Kind | Skill |
| Publisher | [Seth Hobson](/@wshobson) |
| Category | [Data & Analytics](/category/data) |
| Works with | Agent Skills - an open standard supported by 40+ agents |
| Size | 19 KB |
| Licence | MIT |
| Repository | [wshobson/agents](https://github.com/wshobson/agents/tree/b6af3711058190e4b5c5274b9758498fe626ec5a/plugins/dgx-spark-ops/skills/spark-environment-setup) |
| GitHub stars | 39,077 (as of Nov 2026) |
| Installs | 2,922, counted by skills.sh (Nov 2026) |

## Security report

**Verified.** AIR scanned this exact version of Spark Environment Setup and flagged nothing. AIR's verdict: **safe**.

**What Spark Environment Setup's check was run against**

|  |  |
| --- | --- |
| Commit | b6af3711058190e4b5c5274b9758498fe626ec5a |
| Content hash | e59c411f69fad8b448b213db060d334da39e817206a306c0c0cf7bdb1bec7403 |
| Files read | [Browse the files at this exact commit](https://github.com/wshobson/agents/blob/b6af3711058190e4b5c5274b9758498fe626ec5a/plugins/dgx-spark-ops/skills/spark-environment-setup/SKILL.md) |

A scan is named by the exact content it read, not by a version number. A different version is a different artifact, and nothing here applies to a version AIR has not scanned.

[Read the full security report for Spark Environment Setup](/security/spark-environment-setup)

## About Spark Environment Setup

This sets up a working ML training and inference environment on NVIDIA DGX Spark - GB10, aarch64, CUDA 13.

Use it when installing PyTorch, Unsloth, TRL, or vLLM on DGX Spark, when you hit libcudart or wheel-ABI errors on aarch64, or when choosing between NGC containers and bare pip installs.

## What's inside

- SKILL.md - instructions
- references/container-workflow.md - reference
- references/stack-matrix.md - reference

## Questions

### Is Spark Environment Setup safe to install?

AIR scanned this exact version of Spark Environment Setup and flagged nothing. AIR's verdict: safe. The verdict is named by the content that was read, at commit b6af37110581, so it cannot drift onto a version nobody scanned. The full report is at https://airmarket.ai/security/spark-environment-setup.

### How do I install Spark Environment Setup?

Open https://airmarket.ai/app/spark-environment-setup on Air Market and use the install action: Air Market hands Claude a prompt naming this exact skill, and Claude sets it up. Source, pinned to the exact commit this listing describes: https://github.com/wshobson/agents/tree/b6af3711058190e4b5c5274b9758498fe626ec5a/plugins/dgx-spark-ops/skills/spark-environment-setup. Install from that link, not the branch.

## Related on Air Market

- [More Data & Analytics listings](/category/data)
- [Everything by Seth Hobson](/@wshobson)

---

Canonical URL: <https://airmarket.ai/app/spark-environment-setup>