Linux Intel GPU Setup

How AI Playground uses Intel GPUs on Linux and what must already be on the host. AI Playground does not install Linux GPU drivers or kernels — prepare the system before you install the app (AppImage, .deb, or dev build).

TL;DR (host prep before AI Playground)

Follow Intel's OMIX install guide for kernel + compute drivers (intel-omix).

Install Vulkan from Ubuntu (§2b) — required for llama.cpp GPU, not covered by OMIX alone.

Verify Level Zero / Vulkan (§3), then install AI Playground and reinstall backends if needed (§5).

The card showing up in lspci is not enough — that's only the PCI device; you need userspace compute and Vulkan drivers on the host first.

1. How backend → GPU selection works

Each backend independently decides whether to use the GPU on Linux. The decision is made by WebUI/electron/adapters/hardware/deviceDetection.ts, which checks both hardcoded library paths and the dynamic linker cache (ldconfig -p) so it works across distros. The decision is logged to the terminal running npm run dev (not the in-app console, because it runs at service-construction time before the renderer logger attaches):

Linux Level Zero runtime detected — Intel GPU (XPU) enabled
Linux Vulkan loader detected — llama.cpp will use the GPU (ubuntu-vulkan-x64) build

Backend	Variant/build chosen	Gating function	Required runtime
llama.cpp	`ubuntu-vulkan-x64` (GPU) vs `ubuntu-x64` (CPU)	`linuxHasVulkanLoader()`	Vulkan loader + Intel ANV
ComfyUI	`xpu` (torch+xpu) vs `cpu`	`linuxHasLevelZeroRuntime()`	Level Zero loader + Intel L0 GPU
OpenVINO	`GPU` device exposed via OVMS	OpenVINO `ov.Core().available_devices` (Python detection venv)	Level Zero and/or OpenCL

The build/variant is chosen at install time, so after installing the GPU drivers you must reinstall (or re-pick) the affected backends — see §5.

Hardware support note

Upstream PyTorch XPU (torch.xpu, used by ComfyUI XPU) only supports Arc (DG2), Data Center GPU Max, and Core-Ultra Xe iGPUs (Meteor/Lunar/Panther Lake). Older Gen9 iGPUs (e.g. HD Graphics 630) are not supported by XPU — on those, ComfyUI correctly stays on CPU. llama.cpp (Vulkan) and OpenVINO (GPU) can still use Gen9.

2. Host setup before installing AI Playground

Do this on Ubuntu before installing AI Playground. None of these steps are run by the app installer; they are standard system administration.

2a. Intel OMIX (kernel + compute drivers)

Use Intel's documentation as the source of truth for repository setup, kernel requirements, intel-omix installation, render-group access, and upgrades:

Installing Intel Open Middleware Xe

Check your GPU, Ubuntu version, and kernel against Intel's OMIX support matrix before you start. On Ubuntu 24.04, Intel currently documents 24.04.4, 26.04, or 24.04 with the 6.17 HWE kernel for OMIX — follow Intel's prerequisites and installation steps there, not a copy in this repo.

What AI Playground expects after you finish Intel's guide:

Host requirement	Used by
`intel-omix` runtime (Level Zero + OpenCL compute stack)	ComfyUI XPU, OpenVINO GPU
Kernel version Intel documents for your hardware	Stable GPU/NPU binding
Membership in the DRM render group (per Intel's guide)	OpenCL / Level Zero device access

You do not need intel-omix-dev to run AI Playground — that package is for building PyTorch/SYCL workloads locally.

Corporate proxy tip. Intel's guide uses wget for the GPG key. If http(s)_proxy is set but no_proxy includes *.intel.com, wget may hang because repositories.intel.com resolves outside that pattern. Prefer curl -fsSL ${http_proxy:+--proxy "$http_proxy"} https://repositories.intel.com/gpu/intel-graphics.key | sudo gpg --dearmor -o /usr/share/keyrings/intel-graphics.gpg when fetching the key.

Optional: install clinfo from apt if Intel's procedure does not already, so you can verify OpenCL in §3.

2b. Vulkan (llama.cpp GPU build — not in OMIX)

OMIX covers the compute stack for ComfyUI and OpenVINO; llama.cpp still needs Vulkan from Ubuntu. Install this on the host before AI Playground as well:

sudo apt-get install -y libvulkan1 mesa-vulkan-drivers vulkan-tools

libvulkan1 → libvulkan.so.1 loader.
mesa-vulkan-drivers → provides libvulkan_intel.so (the ANV driver that drives Arc/DG2 and Gen9+ iGPUs).

2c. Render-node permissions (if not already done)

Intel's OMIX guide adds your user to the render group. Some setups also need video for legacy i915 paths:

sudo gpasswd -a "$USER" render
sudo gpasswd -a "$USER" video
# log out/in (or `newgrp render`) so the group membership applies

2d. (Optional) Install the Intel NPU userspace driver

Required only if your CPU has an integrated NPU (Meteor/Lunar/Panther/Arrow Lake) and you want OpenVINO to enumerate NPU as an inference target. Without it, Core().available_devices silently returns only ['CPU', 'GPU'] — even though /dev/accel/accel0 is present and intel_vpu is loaded.

The NPU userspace is not in the GPU APT repo above. It ships only as a release tarball on GitHub:

# Pick the latest tag from https://github.com/intel/linux-npu-driver/releases
NPU_TAG=v1.33.0
NPU_BUILD=v1.33.0.20260529-26625960453

cd /tmp
curl -fL ${http_proxy:+--proxy "$http_proxy"} \
  -o linux-npu-driver.tar.gz \
  "https://github.com/intel/linux-npu-driver/releases/download/${NPU_TAG}/linux-npu-driver-${NPU_BUILD}-ubuntu2404.tar.gz"

mkdir -p linux-npu-driver && tar -xzf linux-npu-driver.tar.gz -C linux-npu-driver

sudo apt install -y \
  ./linux-npu-driver/intel-driver-compiler-npu_*.deb \
  ./linux-npu-driver/intel-fw-npu_*.deb \
  ./linux-npu-driver/intel-level-zero-npu_*.deb

Verify (after re-login so render group membership is active):

ls /usr/lib/x86_64-linux-gnu/libze_intel_npu.so*   # plugin must exist
# Use the app's managed Python venv (system python3 won't have openvino):
~/.local/share/ai-playground/resources/OpenVINO/.venv/bin/python3 -c \
  "import openvino as ov; print(ov.Core().available_devices)"
# Expected: ['CPU', 'GPU', 'NPU']

Kernel pairing. The NPU userspace ABI is locked to the kernel intel_vpu module. Panther Lake needs kernel ≥ 6.10; on older Ubuntu kernels install linux-generic-hwe-24.04.

Proxy. GitHub release downloads redirect to AWS, which the *.intel.com no_proxy rule does not cover. The ${http_proxy:+--proxy ...} form above forwards the proxy explicitly when one is set.

2e. Alternative: legacy `unified` repository (hardware outside OMIX)

If your GPU or Ubuntu/kernel combination is not listed in Intel's OMIX support matrix (for example some Arc A-series setups on an older 24.04 kernel), install Level Zero from Intel's generic GPU repo before AI Playground using the legacy path:

curl -fsSL ${http_proxy:+--proxy "$http_proxy"} \
  https://repositories.intel.com/gpu/intel-graphics.key \
  | sudo gpg --dearmor --output /usr/share/keyrings/intel-graphics.gpg

echo "deb [arch=amd64 signed-by=/usr/share/keyrings/intel-graphics.gpg] https://repositories.intel.com/gpu/ubuntu noble unified" \
  | sudo tee /etc/apt/sources.list.d/intel-gpu-noble.list

sudo apt-get update
sudo apt-get install -y libze-intel-gpu1 libze1 intel-opencl-icd clinfo

If apt-get update 404s on unified, Intel may have pinned a dated path (e.g. noble/production/2328 unified). Check https://repositories.intel.com/gpu/ubuntu/dists/ for the current component.

3. Verify the runtime is visible (before / after installing AI Playground)

# Kernel binding + render node (Arc exposes /dev/dri/renderD12x)
ls -l /dev/dri
lspci -nnk -d 8086:5690        # "Kernel driver in use: i915" (or xe) + device id

# Level Zero (ComfyUI XPU + OpenVINO GPU)
ldconfig -p | grep -E 'libze_loader|libze_intel_gpu'
clinfo | grep -iE 'Number of platforms|Device Name'      # expect "Intel(R) Arc(TM) A770"

# Vulkan (llama.cpp)
ldconfig -p | grep libvulkan
vulkaninfo --summary | grep -iE 'deviceName|driverName'  # expect Intel Arc + ANV

If lspci lists the GPU but clinfo / vulkaninfo show no device, the kernel sees the card but the userspace compute/Vulkan driver is missing — finish §2 (OMIX or §2e legacy packages, plus §2b for Vulkan) before installing the app.

Troubleshooting: `clinfo` reports `Number of platforms 0`

Two common causes (often both):

Group membership not applied. After gpasswd -a "$USER" render you must log out and back in (or reboot). Until then your session cannot open /dev/dri/renderD12x, so OpenCL/Level Zero see no device. Verify with groups | tr ' ' '\n' | grep -E 'render|video' and ls -l /dev/dri.
OpenCL ICD not installed. 0 platforms means no ICD is registered — install intel-opencl-icd. Check /etc/OpenCL/vendors/*.icd.

Note the Level Zero loader is separate from the OpenCL ICD: libze_loader.so being present (so linuxHasLevelZeroRuntime() returns true) does not mean the GPU is usable. torch.xpu / OpenVINO also need the Level Zero GPU backend (libze_intel_gpu.so, installed via intel-omix or libze-intel-gpu1) and an accessible render node. Confirm the GPU is actually live (clinfo lists the device, /dev/dri accessible) before reinstalling ComfyUI for XPU, or torch.xpu.device_count() will be 0.

4. Kernel requirements

Use Intel's OMIX guide and support matrix (§2a) as the primary kernel guidance for your GPU. Additional notes for AI Playground backends:

Arc (DG2) needs a kernel with DG2 i915 support. Ubuntu 24.04's 6.8 kernel supports it out of the box when OMIX does not require a newer HWE build. On older kernels you may need i915.force_probe=<device-id> on the kernel command line.
Panther Lake / Lunar Lake / Arrow Lake use the new xe kernel driver (not i915). Needs kernel ≥ 6.10 — install linux-generic-hwe-24.04 on Ubuntu 24.04 if your kernel is older (OMIX may require 6.17 HWE on 24.04; prefer the version Intel documents for your GPU).
Confirm the driver bound with lspci -nnk → Kernel driver in use: i915 (or xe).

4a. Monitoring live GPU usage

For live utilization, frequency, and memory stats on Intel GPUs, use xpu-smi from Intel's XPU Manager (xpumanager) repo. Install and run it on the host per that project's Linux documentation (for example, xpu-smi discovery to list devices, then the stats subcommands while AI Playground is generating).

intel_gpu_top (from intel-gpu-tools) does not support the xe driver yet. If you do not have xpu-smi installed, you can still see which processes hold the render node on xe:

sudo fuser -v /dev/dri/renderD128
# Healthy AI-Playground inference shows: ovms + ai-playground.bin + python (ComfyUI)

For instantaneous frequency on xe without xpu-smi, read sysfs:

cat /sys/class/drm/card0/device/tile0/gt0/freq0/cur_freq      # GPU current MHz
cat /sys/class/drm/card0/device/tile0/gt0/freq0/max_freq

5. After AI Playground is installed — make backends pick up the GPU

Host drivers must already be in place (§2–§3). The GPU build/variant is selected during backend installation inside the app, so after drivers are ready:

ComfyUI — detection flips to xpu and the UI prompts a reinstall on the cpu → xpu variant mismatch. Accept it so torch+xpu wheels are installed.
llama.cpp — CPU and Vulkan share the same standard variant, so it does not auto-reinstall. Uninstall + reinstall llama.cpp so it downloads the ubuntu-vulkan-x64 build. Afterwards --list-devices lists the GPU and --gpu-layers 999 offloads to it.
OpenVINO — restart the backend; the device dropdown will list GPU (detection runs in the OpenVINO Python venv via detect_devices.py).

6. Known follow-up: ComfyUI `comfyui-deps` lock on Linux

After adding sys_platform == 'linux' to comfyui-deps/pyproject.toml and pyproject-flexible-venv.toml, the uv.lock should be regenerated on Linux so the xpu extra resolves to the pytorch-xpu index instead of the default CUDA-bundled PyPI torch. Symptom if stale: uv tries to install nvidia-* / cuda-* packages on a non-NVIDIA Linux host and the env check fails with expected object, received null.

Regenerate on the Linux machine:

cd comfyui-deps
uv lock

7. Relevant source

WebUI/electron/adapters/hardware/deviceDetection.ts — linuxHasLevelZeroRuntime(), linuxHasVulkanLoader() (ldconfig-aware, logged).
WebUI/electron/adapters/backends/comfyUIBackendService.ts — XPU variant selection, oneAPI LD_LIBRARY_PATH, ZE_FLAT_DEVICE_HIERARCHY, lowvram handling.
WebUI/electron/adapters/backends/llamaCppBackendService.ts — Vulkan build selection (resolveDownloadUrl()), --list-devices parsing.
WebUI/electron/adapters/backends/openVINOBackendService.ts — OVMS env (system Python on Linux), Python device-detection venv.
WebUI/electron/adapters/hardware/hardwareDiscovery.ts — lspci-based Intel GPU detection (mode recommendation).

Navigation Menu

Repository navigation

linux-intel-gpu-setup.md

linux-intel-gpu-setup.md

Linux Intel GPU Setup

1. How backend → GPU selection works

Hardware support note

2. Host setup before installing AI Playground

2a. Intel OMIX (kernel + compute drivers)

2b. Vulkan (llama.cpp GPU build — not in OMIX)

2c. Render-node permissions (if not already done)

2d. (Optional) Install the Intel NPU userspace driver

2e. Alternative: legacy `unified` repository (hardware outside OMIX)

3. Verify the runtime is visible (before / after installing AI Playground)

Troubleshooting: `clinfo` reports `Number of platforms 0`

4. Kernel requirements

4a. Monitoring live GPU usage

5. After AI Playground is installed — make backends pick up the GPU

6. Known follow-up: ComfyUI `comfyui-deps` lock on Linux

7. Relevant source

FilesExpand file tree

linux-intel-gpu-setup.md

Latest commit

History

linux-intel-gpu-setup.md

File metadata and controls

Linux Intel GPU Setup

1. How backend → GPU selection works

Hardware support note

2. Host setup before installing AI Playground

2a. Intel OMIX (kernel + compute drivers)

2b. Vulkan (llama.cpp GPU build — not in OMIX)

2c. Render-node permissions (if not already done)

2d. (Optional) Install the Intel NPU userspace driver

2e. Alternative: legacy unified repository (hardware outside OMIX)

3. Verify the runtime is visible (before / after installing AI Playground)

Troubleshooting: clinfo reports Number of platforms 0

4. Kernel requirements

4a. Monitoring live GPU usage

5. After AI Playground is installed — make backends pick up the GPU

6. Known follow-up: ComfyUI comfyui-deps lock on Linux

7. Relevant source

2e. Alternative: legacy `unified` repository (hardware outside OMIX)

Troubleshooting: `clinfo` reports `Number of platforms 0`

6. Known follow-up: ComfyUI `comfyui-deps` lock on Linux