Pro Guides
Essential installation, setup and hardware tutorials for ComfyUI.
Z-Image Turbo Fine-Tune Wars: 4 Checkpoints Compared (ComfyUI)
We tested the 4 most popular Z-Image Turbo fine-tunes with the same prompt and seed. Real finding: 2 of the 4 change pose and gaze, not just skin texture.
Conditioning in ComfyUI: Prompt, Reference and Mask (Krea 2, RTX 3090)
We tested three ways to condition a generation in ComfyUI with Krea 2: prompt, reference (ReferenceLatent) and mask. One did nothing -- pixel diff of zero.
GGUF vs bf16 in ComfyUI: How Much VRAM and Speed You Gain (RTX 3090)
We compared Z-Image Turbo in bf16 against its GGUF Q8_0 quantization in ComfyUI, same prompt and seed. GGUF wins on VRAM, but by far less than the 50% disk savings.
The 9 Krea 2 Style LoRAs: A Real Side-by-Side Comparison (ComfyUI, RTX 3090)
We tested all 9 official Krea 2 style LoRAs with the same prompt, seed and parameters in ComfyUI, with and without their trigger phrase. Includes the real finding: without the trigger, 7 of the 9 lose their specific character and converge to a similar generic style.
GGUF Models in ComfyUI: Run Flux on 8GB VRAM (Quantization Explained)
What GGUF quantization is, how it differs from safetensors, and how to run Flux and other large models on 8GB VRAM GPUs using ComfyUI-GGUF.
How to Install ComfyUI on Windows in 2026 (Portable, Manual & Desktop App)
Install ComfyUI on Windows in under 30 minutes. Desktop installer vs manual Python setup, CUDA/PyTorch matching, and where models go.
How to Install Custom Nodes in ComfyUI (Manager, Git & Fixing Import Failed)
Install ComfyUI custom nodes with ComfyUI Manager or git clone, and fix the most common 'Import Failed' errors after installing one.
How to Use LoRAs in ComfyUI: Loader Node, Strength Settings & Stacking
Add LoRAs to any ComfyUI workflow with the LoraLoader node. Where to download them, strength values for SD/SDXL vs Flux, and how to stack multiple LoRAs.
Krea 2 in ComfyUI: Full Setup Guide for the Turbo Model (Tested on RTX 3090)
How to install and run Krea 2 Turbo in ComfyUI: exact models, a working flat workflow (the official subgraph template is broken), real generation times, and the style LoRAs.
How to Run ComfyUI in the Cloud: RunPod vs Vast.ai Setup Guide (2026)
Run ComfyUI on a rented cloud GPU when your own hardware isn't enough. RunPod vs Vast.ai compared: how each platform works, what to look for, and when cloud beats buying.
Best GPU for ComfyUI in 2026 (Local Build Buyer's Guide)
GPU comparison for ComfyUI and Stable Diffusion in 2026: RTX 3060, 4070, 4090 and AMD. How much VRAM you need for each model and what card to buy on your budget.
How to Reduce VRAM Usage in ComfyUI (Low VRAM / 6GB-8GB Guide)
Every option to reduce VRAM usage in ComfyUI: --lowvram, GGUF, VAE tiling, sequential offload, attention slicing and lightweight models. Guide by GPU tier.
Wan 2.2 vs HunyuanVideo: Which Generates Better Video in ComfyUI
Full Wan 2.2 vs HunyuanVideo comparison in ComfyUI. VRAM, speed, motion quality, prompt understanding, and when to use each one.
What Is ComfyUI? The Complete Beginner's Guide (2026)
Discover what ComfyUI is, how its node system works, and why it's the most powerful local generative AI tool for creating images and video.