SDXL on AMD / Intel GPUs via Vulkan (No CUDA or ROCm)

    1

    Vendor PyTorch builds skip many AMD and Intel cards on Windows and ROCm is Linux-first, so people with a capable 16 GB card are told they cannot run SDXL locally.

    $29

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    SDXL on AMD / Intel GPUs via Vulkan (No CUDA or ROCm)

    Example session with this skill installed

    I have a Radeon RX 6800 XT on Windows and every SDXL guide wants CUDA or ROCm. Get Stable Diffusion XL running locally on this card, fast, without installing a Python stack.

    • Read your context and instructions
    • Compiled the sdxl on amd

    Summary

    SDXL runs on the RX 6800 XT through Vulkan: 8 seconds per 1024x1024 image on the fast preset, 47 seconds on quality, nothing installed system-wide, every downloaded byte hash-verified.

    What was done

    • pwsh -File scripts/Test-Scripts.ps1: 15 checks passed (no download).
    • .\scripts\setup.ps1: plan printed (tool release master-1024-77b72a2, three models, 14.5 GB).
    • .\scripts\setup.ps1 -Install: four files downloaded with resume, each hashed and moved into place only on a match; sd-cli.exe --help answered.
    • .\scripts\generate.ps1 "a red fox in a snowy pine forest at dawn, wildlife photography": Done in 8.3 s: out\20261029-141502-fast-184719-*.png, sidecar JSON beside it.
    • .\scripts\generate.ps1 "an old fisherman mending a net, harbour at dusk" -Preset quality -Width 832 -Height 1216 -Seed 7: Done in 49.1 s.

    Why it was slow before

    The card fails the direct-convolution self-check, so the tool decoded each image on the slow path: 34 s per image. --conv-direct auto,vae=on (the default in generate.ps1) decodes in 1.5 s and matched the default decode at 50.7 dB PSNR. Flash attention was measured 7x slower on this card and stays off.

    Licences

    SDXL base and Lightning: CreativeML OpenRAIL++-M (commercial use allowed with use restrictions). Tool and VAE: MIT. Face-identity adapters left out: their face models are non-commercial.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Vendor PyTorch builds skip many AMD and Intel cards on Windows and ROCm is Linux-first, so people with a capable 16 GB card are told they cannot run SDXL locally. They can: stable-diffusion.cpp over Vulkan runs it on anything with a Vulkan driver. This kit is the pinned install, the presets and the measurements that make it fast instead of painfully slow.

    setup.ps1: plan mode prints what will be downloaded and where; install mode fetches the pinned stable-diffusion.cpp Vulkan build and three models (SDXL base 1.0, SDXL-Lightning 4-step, the fp16-fix VAE; 14.5 GB) with resume, hashes each .part file and moves it into place only when it matches the approved SHA-256, deletes a mismatch, and re-runs download nothing already correct. Nothing system-wide, no driver changes.

    generate.ps1: fast (Lightning 4-step, guidance 1, about 8 s at 1024x1024, 6 s each in a batch) and quality (SDXL base, 30 steps, guidance 7, about 47 s); size validation (multiples of 64, 512 to 1536); seed, count, negative prompt; a .json sidecar beside every PNG; -DryRun prints the exact command.

    Why it is fast (reference/speed-settings-and-why.md, measured on an RX 6800 XT): the card fails the direct-convolution self-check so decode took 34 s; forcing it on for the VAE only decodes in 1.5 s and matched the default at 50.7 dB PSNR, while tiled decode (21 dB) and CPU decode (60 s) were rejected; flash attention measured 7x slower; the Lightning LoRA matched 0 of 2,364 tensors so the full checkpoint is used; the fp16-fix VAE avoids black images.

    Licences (reference/models-and-licences.md): every file with source, revision, licence and hash; OpenRAIL++-M use restrictions; why face adapters are left out. 15 offline checks (Test-Scripts.ps1) that download nothing. PowerShell 5.1 or 7.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Video Skills.

    Requires GPU with 8GB+ VRAM and Vulkan support. Windows only. No face-identity adapters included. Downloads are 14.5 GB. One generation job at a time to avoid VRAM exhaustion.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 19 days ago

    Windows only

    Software
    PowerShell 5.1 or 7curl.exeGPU with Vulkan and 8 GB+ VRAM

    16 GB of VRAM is comfortable.

    Permissions

    Network Access
    Read Files
    Terminal / Shell
    Write Files

    Shell to run the PowerShell scripts and tests; read/write files for the tool, models and output folders. Network access is used only by setup.ps1 -Install to download from github.com and huggingface.co; generation makes no network calls. No environment variables.

    Listed19 days ago

    What's inside

    Frequently Asked Questions

    This skill is designed for Windows users with AMD or Intel GPUs who cannot use CUDA-based tools. It provides a portable, Vulkan-based implementation of SDXL that works without needing to install Python, Git, or complex driver toolchains like ROCm.

    You need a Windows 10 or 11 system with a GPU supporting Vulkan and at least 8GB of VRAM (16GB is recommended). No administrator rights are required, as the skill uses portable binaries and hash-verified downloads that don't touch your system folders.

    You receive a suite of production-ready PowerShell scripts that handle environment validation, automated installation of optimized models, and pre-configured generation modes for both speed and quality. It also includes performance-tuning logic that significantly reduces image decoding times on Vulkan hardware.

    No, this is a "zero-install" solution. All necessary components, including the pinned version of stable-diffusion.cpp and the specific GGUF models, are downloaded into a local folder via the script, keeping your main operating system clean and dependency-free.

    The skill includes a 'Validation' phase that performs 15 offline checks to ensure your environment is compatible before downloading any files. If a download is interrupted, the scripts feature automatic resume support and cryptographic hash-verification to ensure your 14.5 GB model files are not corrupted.

    Currently, this skill focuses on high-quality text-to-image generation using SDXL Base and Lightning models. It does not support additional plugins like ControlNet or LoRAs, as it is optimized for stable, reproducible performance across diverse Vulkan-compatible hardware.

    New in Image & Video