Renderer & Engine Selection

VeloxKit has three rendering backends with very different memory profiles. Understanding how the selection works lets you ship the right binary for your audience.

The numbers

Measured on real apps (2026-05, hello-world / Calculator):

The backend is the dominant cost. The jump from TinySkia to Vello is ~188 MB on hello-world — larger than Electron's entire binary.

How auto works

auto runs after the GPU adapter is created and classifies the hardware into one of four tiers using wgpu::DeviceType and the PCI vendor ID:

No GPU / WARP / llvmpipe                         →  TinySkia
Integrated GPU (Intel UHD, iGPU, Apple Silicon)  →  TinySkia
Intel Arc discrete GPU (vendor 0x8086, discrete)  →  FemtoVG
NVIDIA or AMD discrete GPU                         →  Vello

The decision is logged at startup so it's always visible:

[velox] renderMode=auto → skia    (Intel(R) UHD Graphics 770, IntegratedGpu)
[velox] renderMode=auto → femtovg (Intel(R) Arc A770 Graphics, DiscreteGpu)
[velox] renderMode=auto → vello   (NVIDIA GeForce RTX 4080, DiscreteGpu)

VELOX_CPU_RENDER=1 overrides all tiers and forces TinySkia regardless of hardware.

App persona recommendations

Use this as a starting point when choosing explicitly:

Explicit config

Override auto when you know your target hardware:

// veloxkit.config.ts
export default defineConfig({
  window: {
    renderMode: 'skia',     // always TinySkia — minimum footprint
    // renderMode: 'femtovg',  // always FemtoVG — GPU animations
    // renderMode: 'gpu',      // always Vello   — max quality
    // renderMode: 'auto',     // heuristic (default)
  },
})

Force TinySkia at runtime (useful for CI or debugging):

VELOX_CPU_RENDER=1 veloxkit dev
VELOX_CPU_RENDER=1 veloxkit build

Fallback chain

If a backend fails to initialise, VeloxKit falls back automatically:

vello → femtovg → skia → ERROR
femtovg → skia → ERROR
skia → ERROR (cannot render)

The fallback is logged so it's never silent.

Edge cases

Hybrid GPU laptops (switchable graphics)

Laptops with both an iGPU and a dGPU let the OS decide which adapter wgpu sees. With PowerPreference::HighPerformance, wgpu usually gets the dGPU. If the OS pins the iGPU, auto will pick TinySkia.

Vulkan on iGPU (Linux)

The Velox GPU adapter probe prefers DX12 on Windows to avoid the Vulkan allocator pre-reserving large pools on iGPU hardware. On Linux with Vulkan + iGPU, auto picks TinySkia which bypasses this entirely.

macOS (Apple Silicon)

Apple Silicon reports as IntegratedGpu (Metal). auto picks TinySkia. Set renderMode: 'femtovg' or 'gpu' explicitly if you want GPU rendering on macOS.

VMs and CI

wgpu::DeviceType::VirtualGpu (VMware, Hyper-V, VirtualBox with GPU passthrough) → TinySkia. Software adapters (no GPU at all) → TinySkia.

Keeping the numbers honest

After each release, re-measure on hello-world and calculator and update this guide if the numbers change. The matrix above reflects measurements from 2026-05. If you add heavy assets or large dependency trees, your app's footprint will differ — use veloxkit dev --perf to see live RSS.