System Requirements

Hardware and software requirements for running 360 Hextile

Last updated: 2026-08-24

System Requirements

Hardware Requirements

When available, the System Requirements card GPU picker on the marketing site is the source of truth for named-GPU fit against the current packaged app runtime; the table below remains broad historical guidance.

Pick your card in the picker on the pricing or features page and the chart measures every pipeline's VRAM requirement against your card's memory. Anything above the line runs today; anything below it needs a larger card. Cards whose architecture the packaged runtime does not yet support are labelled as such rather than shown as out of memory.

GPU (REQUIRED)

  • CUDA-capable NVIDIA GPU with a current CUDA-compatible driver (CUDA 12.1+)
  • SDXL (default) needs 12 GB VRAM. Lighter pipelines can run with less.
  • Recommended: 16 GB VRAM
  • Optimal: 24 GB VRAM

GPU Compatibility Table

GPU VRAM Recommended Processors Performance
RTX 4090 24GB All processors except 40+ GB precision variants Excellent
RTX 4080 16GB Most processors; excludes Chroma1-HD and 40+ GB precision variants Excellent
RTX 4070 Ti 12GB SDXL, TeleStyle Alpha, Z-Image Turbo, SD 1.5/2.1 Great
RTX 4070 12GB SDXL, TeleStyle Alpha, Z-Image Turbo, SD 1.5/2.1 Great
RTX 4060 Ti 8GB SD 1.5, SD 2.1 Good
RTX 4060 8GB SD 1.5, SD 2.1 Good
RTX 3090 24GB All processors Excellent
RTX 3080 10-12GB SD 1.5/2.1; 12 GB models also support SDXL Great
RTX 3070 8GB SD 1.5, SD 2.1 Good
RTX 3060 12GB SDXL, TeleStyle Alpha, Z-Image Turbo, SD 1.5/2.1 Good
RTX 2080 Ti 11GB SD 1.5, SD 2.1 Good
RTX 2070/2060 6-8GB SD 1.5, SD 2.1 Adequate

VRAM Usage by Processor

Processor Minimum VRAM Recommended VRAM
SD 1.5 Inpaint 6 GB 8 GB
SD 2.1 Inpaint 8 GB 10 GB
SDXL Inpaint 12 GB 14 GB
SD 3.5 Inpaint 16 GB 18 GB
FLUX.1 Schnell 24 GB 24 GB; BF16-capable NVIDIA CUDA GPU required
FLUX.2 Klein 13 GB 16 GB
Qwen-Image Alpha (GGUF Q4) 13 GB 16 GB
Qwen-Image Alpha (BF16) 40 GB 48 GB
TeleStyle Alpha (GGUF Q4) 12 GB 16 GB
Chroma1-HD Alpha 19 GB 24 GB
Z-Image Turbo 12 GB 14 GB
Real-ESRGAN Alpha 4 GB 6 GB
Coverage Test Alpha 0 GB 0 GB
Post Process 0 GB 0 GB
Format Converter Alpha 0 GB 0 GB

Note: Only ONE diffusion model is loaded in VRAM at a time. Switching processors auto-unloads the previous model.

System RAM

  • Minimum: 16 GB
  • Recommended: 32 GB

Storage

  • Current Windows installer: about 2.9 GB; it unpacks to about 6.2 GB
  • Default first-run SDXL download: about 14 GB. Other pipelines are typically about 10–20 GB; some optional precision variants are larger.
  • Render output: varies by resolution and project
  • Recommended: 50 GB free SSD space

CPU

  • Modern multi-core (Intel i5 / AMD Ryzen 5 or better)
  • CPU is not the bottleneck — GPU handles all AI processing

Software Requirements

Operating System

  • Windows only: Windows 10 (64-bit, 21H2+) or Windows 11
  • macOS is NOT supported (requires NVIDIA CUDA GPU)

NVIDIA Driver

  • Use a current CUDA-compatible NVIDIA driver (CUDA 12.1+)
  • The latest Game Ready or Studio driver is recommended
  • Check your driver: run nvidia-smi in Command Prompt

WebView2 Runtime

  • Pre-installed on Windows 10/11
  • If missing: download from Microsoft

Not Supported

  • AMD GPUs (no CUDA support)
  • Intel Arc GPUs
  • Apple Silicon Macs
  • Integrated graphics (Intel HD, AMD Vega)
  • Cloud GPU instances (app requires local display)

Checking Your System

  • In-app: Settings → GPU Information panel shows your GPU model, VRAM, and driver version
  • Command line: nvidia-smi shows GPU model, VRAM, driver version
  • GPU-Z (free tool) for detailed GPU specs

See also: Pipelines for processor details | Settings for the GPU information panel | Model Management for model download sizes

Esc
Searching...
No results found.
Type to search the documentation
Navigate Select Esc Close

We use cookies for analytics to understand how visitors use our site. Privacy Policy