Integrated GPU, unbeatable price
The GPU is built into the CPU (iGPU): no dedicated card, lower power draw, hardware acceleration from €9.99/month.
iGPU / APU servers · from €9.99/mo
Integrated-GPU (iGPU / APU) servers, for real compute without breaking the bank.
Every iGPU plan runs on dedicated hardware with an Intel GPU: the integrated Intel UHD 770, or a dedicated Intel HD Graphics 530 on iGPU Low cost, exposed through OpenCL, Vulkan and VAAPI. Great for transcoding, AI inference, headless rendering and scientific computing. k3s / Kubernetes ready.
- Intel UHD 770
- k3s / k8s ready
- OpenCL · Vulkan · VAAPI
- Root SSH · IPv6 · Docker
Real hardware acceleration
OpenCL and Vulkan for compute, VAAPI / Intel QuickSync for video, OpenVINO for inference. Drivers ready.
Cluster-ready
Our iGPU plans are k3s and Kubernetes compatible: the GPU (/dev/dri) can be used by your pods, e.g. with the Intel GPU device plugin.
GPU VPS pricing — all iGPU / APU
From a cheap entry-level GPU VPS to a cluster: every plan ships an integrated GPU and is k3s / Kubernetes ready.
iGPU Low cost
€9.99/mo excl. VAT
- Dedicated Intel HD Graphics 530 GPU
- 1 vCPU, 4 GB RAM, 50 GB HDD
- OpenCL · Vulkan · VAAPI
- Basic transcoding
- k3s/k8s ready · root SSH · IPv6 · Docker
- Support response within 2 business days
Need more storage? Add it to your server:
- 1 TB redundant network storage€9.99/mo excl. VAT
iGPU Start
€14.99/mo excl. VAT
- Shared Intel UHD 770 iGPU* · Perf 1
- 2 vCPU, 8 GB RAM, 60 GB SSD
- OpenCL · Vulkan · VAAPI
- Inference · rendering · transcoding
- k3s/k8s ready · root SSH · IPv6 · Docker
- Support response within 2 business days
Need more storage? Add it to your server:
- +60 GB NVMe SSD€5/mo excl. VAT
- +120 GB NVMe SSD€9/mo excl. VAT
- 1 TB redundant network storage€9.99/mo excl. VAT
iGPU Power
€29.99/mo excl. VAT
- Shared Intel UHD 770 iGPU* · Perf 2
- 4 vCPU, 16 GB RAM, 120 GB SSD
- OpenCL · Vulkan · VAAPI
- Inference · rendering · transcoding
- k3s/k8s ready · root SSH · IPv6 · Docker
- Support response within 1 business day
Need more storage? Add it to your server:
- +60 GB NVMe SSD€5/mo excl. VAT
- +120 GB NVMe SSD€9/mo excl. VAT
- 1 TB redundant network storage€9.99/mo excl. VAT
iGPU Pro
€44.99/mo excl. VAT
- Shared Intel UHD 770 iGPU* · Perf 3
- 6 vCPU, 24 GB RAM, 240 GB SSD
- OpenCL · Vulkan · VAAPI
- Inference · rendering · transcoding · AI
- k3s/k8s ready · root SSH · IPv6 · Docker
- Support response within 24 hours, including weekends and public holidays
Need more storage? Add it to your server:
- +60 GB NVMe SSD€5/mo excl. VAT
- +120 GB NVMe SSD€9/mo excl. VAT
- 1 TB redundant network storage€9.99/mo excl. VAT
Dedicated NVIDIA GPU servers
A physical graphics card dedicated to your VM, for 3D rendering, games and video encoding.
NVIDIA GeForce GT 620M
€14.99/mo excl. VAT
- Dedicated GPU · 1 GB DDR3 VRAM
- Dedicated CPU: Intel Core i5-4210U at 1.70 GHz · low-power Haswell · 2 cores / 4 threads
- 4 GB RAM · 100 GB HDD
NVIDIA GeForce GTX 1060
€49.99/mo excl. VAT
- Dedicated GPU · 6 GB VRAM
- AMD Ryzen 5 2600 host CPU · 4 vCPUs assigned to the VM
- 8 GB RAM · 100 GB SSD
iGPU plans: integrated GPU (OpenCL · Vulkan · VAAPI), k3s / Kubernetes ready, root SSH, dedicated IPv6 included (dedicated IPv4 as an option: +€5/month). Options: k3s cluster +€5/node, managed node +€5/node, Windows 11 +€5/month, daily backup (7-day retention) +€2/month, managed Jellyfin (turnkey media server, GPU transcoding on) +€9.99/month. Two dedicated NVIDIA GPU plans are also available above.
Mac — macOS in the cloud
A dedicated physical Mac: SSH + VNC access, IPv4 and IPv6 included. Great for iOS/macOS builds, Xcode and Apple CI.
Mac mini M4 · 256 GB
€95/mo excl. VAT
- Apple M4 · 10-core CPU / 10-core GPU
- 16 GB RAM · 256 GB SSD
- macOS · SSH + VNC access
- Dedicated physical Mac · IPv4 + IPv6 included
Mac mini M4 · 512 GB
€109/mo excl. VAT
- Apple M4 · 10-core CPU / 10-core GPU
- 16 GB RAM · 512 GB SSD
- macOS · SSH + VNC access
- Dedicated physical Mac · IPv4 + IPv6 included
Mac mini M4 · 24 GB RAM
€139/mo excl. VAT
- Apple M4 · 10-core CPU / 10-core GPU
- 24 GB RAM · 512 GB SSD
- macOS · SSH + VNC access
- Dedicated physical Mac · IPv4 + IPv6 included
Mac mini M6 · 256 GB
€159/mo excl. VAT
- Apple M6 · 12-core CPU / 12-core GPU
- 16 GB RAM · 256 GB SSD
- macOS · SSH + VNC access
- Dedicated physical Mac · IPv4 + IPv6 included
Mac mini M6 · 512 GB
€199/mo excl. VAT
- Apple M6 · 12-core CPU / 12-core GPU
- 16 GB RAM · 512 GB SSD
- macOS · SSH + VNC access
- Dedicated physical Mac · IPv4 + IPv6 included
Mac mini M6 · 24 GB RAM
€229/mo excl. VAT
- Apple M6 · 12-core CPU / 12-core GPU
- 24 GB RAM · 512 GB SSD
- macOS · SSH + VNC access
- Dedicated physical Mac · IPv4 + IPv6 included
Mac mini M4 Pro · 512 GB
€239/mo excl. VAT
- Apple M4 Pro · 12-core CPU / 16-core GPU
- 24 GB RAM · 512 GB SSD
- macOS · SSH + VNC access
- Dedicated physical Mac · IPv4 + IPv6 included
Mac mini M4 Pro · 1 TB
€309/mo excl. VAT
- Apple M4 Pro · 12-core CPU / 16-core GPU
- 48 GB RAM · 1 TB SSD
- macOS · SSH + VNC access
- Dedicated physical Mac · IPv4 + IPv6 included
Mac mini M4 Pro · 4 TB
€459/mo excl. VAT
- Apple M4 Pro · 14-core CPU / 20-core GPU
- 64 GB RAM · 4 TB SSD
- macOS · SSH + VNC access
- Dedicated physical Mac · IPv4 + IPv6 included
Mac Pro M2 Ultra
€909/mo excl. VAT
- Apple M2 Ultra · 24-core CPU / 60-core GPU
- 64 GB RAM · 1 TB SSD
- macOS · SSH + VNC access
- Dedicated physical Mac · IPv4 + IPv6 included
iGPU / APU: the integrated GPU, on every iGPU plan
An iGPU (or APU) is a GPU built into the processor: shared memory, low power, and enough performance for a large share of GPU workloads — for a fraction of a dedicated GPU. Our servers use the Intel UHD 770.
On every iGPU plan
Every iGPU plan (Start, Power, Pro) includes the Intel UHD 770 GPU, exposed to the OS (/dev/dri), drivers installed.
Standard APIs
OpenCL and Vulkan compute, VAAPI hardware encode (Intel QuickSync), OpenVINO inference. Docker friendly.
Native k3s / k8s
k3s / Kubernetes compatible: the GPU (/dev/dri) can be used by your pods, e.g. with the Intel GPU device plugin.
Power efficient
No 250 W dedicated GPU: integrated acceleration stays efficient, ideal for 24/7 multi-node clusters.
iGPU / APU use cases
What you can run today on an integrated GPU, from a single box to a cluster.
Julia & scientific computing
Julia jobs and notebooks accelerated on iGPU via OpenCL: algebra, simulation, data.
Learn more →Xvfb & headless rendering
Virtual framebuffer for UI tests, headless browsers (Selenium/Playwright) and server OpenGL rendering.
Learn more →Video transcoding
Hardware encode VAAPI (Intel QuickSync) with FFmpeg: streaming, thumbnails, media pipelines.
Learn more →AI inference
Small local LLMs (Gemma, Phi, Qwen) from Hugging Face via Ollama / llama.cpp, plus vision via OpenVINO — simple, cheap AI on the iGPU.
Learn more →3D & OpenGL/Vulkan rendering
Headless Blender (EEVEE), OpenGL/Vulkan engines, server-side image generation.
Learn more →k3s / k8s clusters
Containerized GPU workloads in HA: multiple iGPU nodes, GPU scheduling, GitOps.
Learn more →Selenium & Playwright GPU
Selenium/Playwright browser tests accelerated by the integrated GPU: WebGL, Canvas and video, in CI.
Learn more →FAQ (iGPU / APU, k3s)
What is an iGPU / APU server?
A server whose GPU is built into the processor. Our servers use the Intel UHD 770 iGPU: hardware acceleration (OpenCL, Vulkan, VAAPI) with no dedicated card.
Which plans include the integrated GPU?
Every iGPU plan (Start, Power, Pro) includes the Intel UHD 770 GPU, exposed to the OS (/dev/dri), drivers installed, usable natively or in Docker containers.
Can I use k3s or Kubernetes with the GPU?
Yes, the iGPU plans are k3s/k8s compatible: the GPU (/dev/dri) can be used by your pods, e.g. with the Intel GPU device plugin. k3s cluster option at +€5/node.
Which GPU workloads are possible?
Transcoding (VAAPI/QuickSync), inference (OpenVINO), compute (Julia, OpenCL), headless rendering (Xvfb, Blender), UI tests, CI/CD.
How much does it cost?
From €9.99/month for iGPU Low cost to €44.99/month for iGPU Pro. k3s and managed-node options at €5/node.
Can I have Windows 11 instead of Linux?
Yes, Windows 11 is available as an option for an extra €5/month. By default our servers run Linux (integrated GPU exposed via /dev/dri).
How do I start?
Reserve in the shop by creating your account, or email contact@linuxgpu.eu. Root SSH access within minutes.
Latest from the blog
Practical guides to run your GPU workloads on an integrated GPU — without renting a €300-500/month dedicated GPU.
Intel Quick Sync (VAAPI) on a Linux VPS: ffmpeg commands that work
Copy-paste ffmpeg VAAPI commands for an Intel UHD 770 on Linux: vainfo, H.264/HEVC encode (classic and low_power), 10-bit HEVC, GPU transcode with scale_vaapi, AV1 decode, GPU monitoring.
Read the article →Shared Intel UHD 770 under full load: a 20-minute GPU stress test
Several servers loading the same shared Intel UHD 770 at once for 20 minutes: H.264, 4K HEVC and AV1 transcodes plus Vulkan/OpenCL. 0 failures, 0 GPU hangs, measured throughput per server.
Read the article →GPU Cloud: VM or Container, Dedicated or Shared GPU? What Price Comparators Don't Show
GPU price comparators list model, VRAM and price per hour, not whether you get a VM or a container, a whole GPU or a share. Live marketplace data, commands to check from inside, buyer checklist.
Read the article →Start your GPU compute today.
Contact us for your iGPU server or your k3s cluster.
contact@linuxgpu.eu