If VRAM is larger than system RAM, all packages using the GPU cause ERR

Nobody has claimed this yet.

[WslLogs-2026-05-31_22-03-33.tar.gz](https://github.com/user-attachments/files/28437987/WslLogs-2026-05-31_22-03-33.tar.gz)

### Windows Version

Microsoft Windows [Version 10.0.26100.8457]

### WSL Version

2.7.3.0

### Are you using WSL 1 or WSL 2?

- [x] WSL 2
- [ ] WSL 1

### Kernel Version

Linux version 6.6.114.1-microsoft-standard-WSL2

### Distro Version

Ubuntu-24.04

### Other Software

In WSL,

Docker version 29.5.2, build 79eb04c
NVIDIA-SMI 595.71.01 Driver Version: 596.36 CUDA Version: 13.2

~$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2023 NVIDIA Corporation
Built on Fri_Jan__6_16:45:21_PST_2023
Cuda compilation tools, release 12.0, V12.0.140
Build cuda_12.0.r12.0/compiler.32267302_0

### Repro Steps

My system has 64GB RAM and RTX PRO 6000 96GB VRAM.

1. In Powershell,
wsl --install Ubuntu-24.04
2. in WSL
```
# install docker, Nvidia-container-toolkit, cuda tollkit
# https://docs.docker.com/engine/install/ubuntu/
# Add Docker's official GPG key:
sudo apt update
sudo apt install ca-certificates curl
sudo install -m 0755 -d /etc/apt/keyrings
sudo curl -fsSL https://download.docker.com/linux/ubuntu/gpg -o /etc/apt/keyrings/docker.asc
sudo chmod a+r /etc/apt/keyrings/docker.asc

# Add the repository to Apt sources:
sudo tee /etc/apt/sources.list.d/docker.sources <

Start by reviewing the attached WslLogs-2026-05-31_22-03-33.tar.gz and reproducing the listed Docker command on WSL 2 with Ubuntu-24.04 and the reported GPU and RAM configuration. Done means loading the model no longer crashes or freezes the Windows host, with the container instead completing normally or reporting an OOM.

Written by the indexing model from the issue text.