---
title: "Vision-Language Model Fine-tuning — Troubleshooting"
canonical: "https://build.nvidia.com/spark/vlm-finetuning/troubleshooting.md"
---

> [!NOTE]
> DGX Spark uses a Unified Memory Architecture (UMA), which enables dynamic memory sharing between the GPU and CPU. 
> With many applications still updating to take advantage of UMA, you may encounter memory issues even when within 
> the memory capacity of DGX Spark. If that happens, manually flush the buffer cache with:
```bash
sudo sh -c 'sync; echo 3 > /proc/sys/vm/drop_caches'
```