cd /news/ai-infrastructure/nbd-vram-provides-swap-space-on-your… · home topics ai-infrastructure article
[ARTICLE · art-19499] src=phoronix.com pub= topic=ai-infrastructure verified=true sentiment=· neutral

NBD-VRAM Provides Swap Space On Your NVIDIA GeForce GPUs

An open-source developer released NBD-VRAM, a tool that creates swap space on consumer NVIDIA GeForce GPUs under Linux. The software targets laptop users with soldered memory who need additional system memory by carving out swap space from the GPU's vRAM via the NVIDIA CUDA driver API and Linux Network Block Device protocol. NBD-VRAM requires the official NVIDIA Linux graphics driver stack and is designed for consumer GPUs where the NVIDIA P2P API does not work.

read2 min publishedJun 1, 2026

An open-source developer has created NBD-VRAM as a way to create swap space on your consumer NVIDIA GPU's video memory under Linux.

Advertised primarily for those with laptops having soldered memory and needing more system memory space while also having a consumer NVIDIA GeForce RTX GPU in the laptop, NBD-VRAM provides a pathway to carving out some swap space from that vRAM.

NBD-VRAM was published yesterday and consists of a small daemon to allocate vRAM via the NVIDIA CUDA driver API and then it is exposed using the Linux Network Block Device (NBD) protocol on a Unix socket. Ultimately this path exposes a portion of the NVIDIA GPU vRAM as a conventional Linux swap device.

NBD-VRAM is designed explicitly to work with NVIDIA's consumer GPUs where the NVIDIA P2P API and alternatives do not work. You do need to be running the official NVIDIA Linux graphics driver stack for CUDA support as opposed to using Nouveau/Nova.

Those wishing to learn more about this MIT-licensed NBD-VRAM open-source project can do so via

Advertised primarily for those with laptops having soldered memory and needing more system memory space while also having a consumer NVIDIA GeForce RTX GPU in the laptop, NBD-VRAM provides a pathway to carving out some swap space from that vRAM.

NBD-VRAM was published yesterday and consists of a small daemon to allocate vRAM via the NVIDIA CUDA driver API and then it is exposed using the Linux Network Block Device (NBD) protocol on a Unix socket. Ultimately this path exposes a portion of the NVIDIA GPU vRAM as a conventional Linux swap device.

NBD-VRAM is designed explicitly to work with NVIDIA's consumer GPUs where the NVIDIA P2P API and alternatives do not work. You do need to be running the official NVIDIA Linux graphics driver stack for CUDA support as opposed to using Nouveau/Nova.

Those wishing to learn more about this MIT-licensed NBD-VRAM open-source project can do so via

── more in #ai-infrastructure 4 stories · sorted by recency
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/nbd-vram-provides-sw…] indexed:0 read:2min 2026-06-01 ·