Learn to master ComfyUI with step-by-step tutorials, downloadable workflows and technical guides.https://comfylab.dev/en-ushttps://comfylab.dev/blog/errors/comfyui-import-failed-custom-node-errors/https://comfylab.dev/blog/errors/comfyui-import-failed-custom-node-errors/Fix ModuleNotFoundError and 'Import Failed' errors in ComfyUI custom nodes. Install missing dependencies, resolve version conflicts, and fix Windows C++ build failures.Fri, 03 Jul 2026 00:00:00 GMTsavienhttps://comfylab.dev/blog/errors/cuda-out-of-memory-comfyui/https://comfylab.dev/blog/errors/cuda-out-of-memory-comfyui/Fix the CUDA out of memory error in ComfyUI step by step. Memory flags, resolution, quantized models and VAE tiling for low-VRAM GPUs (4GB, 6GB, 8GB).Thu, 02 Jul 2026 00:00:00 GMTsavienhttps://comfylab.dev/blog/errors/torch-not-compiled-with-cuda-comfyui/https://comfylab.dev/blog/errors/torch-not-compiled-with-cuda-comfyui/Fix the 'Torch is not compiled with CUDA support' error in ComfyUI. Real cause: CPU-only PyTorch, wrong CUDA index, or a broken virtual environment.Fri, 03 Jul 2026 00:00:00 GMTsavienhttps://comfylab.dev/blog/guides-pro/best-gpu-for-comfyui/https://comfylab.dev/blog/guides-pro/best-gpu-for-comfyui/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.Thu, 02 Jul 2026 00:00:00 GMTsavienhttps://comfylab.dev/blog/guides-pro/gguf-models-comfyui-run-flux-8gb-vram/https://comfylab.dev/blog/guides-pro/gguf-models-comfyui-run-flux-8gb-vram/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.Fri, 03 Jul 2026 00:00:00 GMTsavienhttps://comfylab.dev/blog/guides-pro/how-to-use-loras-comfyui/https://comfylab.dev/blog/guides-pro/how-to-use-loras-comfyui/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.Fri, 03 Jul 2026 00:00:00 GMTsavienhttps://comfylab.dev/blog/guides-pro/krea-2-comfyui-guide-turbo-model/https://comfylab.dev/blog/guides-pro/krea-2-comfyui-guide-turbo-model/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.Fri, 03 Jul 2026 00:00:00 GMTsavienhttps://comfylab.dev/blog/guides-pro/how-to-install-custom-nodes-comfyui/https://comfylab.dev/blog/guides-pro/how-to-install-custom-nodes-comfyui/Install ComfyUI custom nodes with ComfyUI Manager or git clone, and fix the most common 'Import Failed' errors after installing one.Fri, 03 Jul 2026 00:00:00 GMTsavienhttps://comfylab.dev/blog/guides-pro/reduce-vram-usage-comfyui/https://comfylab.dev/blog/guides-pro/reduce-vram-usage-comfyui/Every option to reduce VRAM usage in ComfyUI: --lowvram, GGUF, VAE tiling, sequential offload, attention slicing and lightweight models. Guide by GPU tier.Thu, 02 Jul 2026 00:00:00 GMTsavienhttps://comfylab.dev/blog/guides-pro/how-to-install-comfyui-windows/https://comfylab.dev/blog/guides-pro/how-to-install-comfyui-windows/Install ComfyUI on Windows in under 30 minutes. Desktop installer vs manual Python setup, CUDA/PyTorch matching, and where models go.Fri, 03 Jul 2026 00:00:00 GMTsavienhttps://comfylab.dev/blog/guides-pro/runpod-vs-vastai-comfyui-cloud-setup/https://comfylab.dev/blog/guides-pro/runpod-vs-vastai-comfyui-cloud-setup/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.Fri, 03 Jul 2026 00:00:00 GMTsavienhttps://comfylab.dev/blog/nodes/ksampler-explained-comfyui-steps-cfg-samplers-schedulers/https://comfylab.dev/blog/nodes/ksampler-explained-comfyui-steps-cfg-samplers-schedulers/What the KSampler node actually does in ComfyUI, and how to set steps, CFG scale, sampler and scheduler to control quality, speed and style.Fri, 03 Jul 2026 00:00:00 GMTsavienhttps://comfylab.dev/blog/workflows/comfyui-upscale-workflow-ultimate-sd-upscale/https://comfylab.dev/blog/workflows/comfyui-upscale-workflow-ultimate-sd-upscale/Upscale ComfyUI images to 4K without running out of VRAM. Ultimate SD Upscale, ESRGAN models, and tiled VAE decode explained -- with a downloadable workflow.Fri, 03 Jul 2026 00:00:00 GMTsavienhttps://comfylab.dev/blog/guides-pro/what-is-comfyui/https://comfylab.dev/blog/guides-pro/what-is-comfyui/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.Thu, 02 Jul 2026 00:00:00 GMTsavienhttps://comfylab.dev/blog/workflows/local-ai-video-models-2026-real-comparison-rtx-3090/https://comfylab.dev/blog/workflows/local-ai-video-models-2026-real-comparison-rtx-3090/LTXV-2.3 distilled, LTXV-2.3 dev, SCAIL-2, Wan 2.1 I2V, and Wan 2.2 I2V, all tested for real on the same hardware. Real timings, real crashes, real quality differences -- not a spec-sheet comparison.Fri, 03 Jul 2026 00:00:00 GMTsavienhttps://comfylab.dev/blog/workflows/ltxv-2-3-rtx-video-super-resolution-comfyui-workflow/https://comfylab.dev/blog/workflows/ltxv-2-3-rtx-video-super-resolution-comfyui-workflow/Step-by-step reproduction of a YouTube LTXV-2.3 text-to-video-with-audio workflow chained into Nvidia's RTX Video Super Resolution node, with real generation times, VRAM usage, and the gotchas I hit on an RTX 3090.Fri, 03 Jul 2026 00:00:00 GMTsavienhttps://comfylab.dev/blog/workflows/ltxv-2-3-dev-model-oom-rtx-3090-gguf-bug/https://comfylab.dev/blog/workflows/ltxv-2-3-dev-model-oom-rtx-3090-gguf-bug/A real, reproducible OOM crash reproduced 4/4 times running LTXV-2.3's non-distilled dev model on a 24GB RTX 3090: the actual bottleneck wasn't VRAM, it was system RAM, and the GGUF loader path is the common factor.Fri, 03 Jul 2026 00:00:00 GMTsavienhttps://comfylab.dev/blog/workflows/scail-2-character-replacement-comfyui-workflow/https://comfylab.dev/blog/workflows/scail-2-character-replacement-comfyui-workflow/Tested Comfy-Org's official SCAIL-2 template for AI character replacement: swap a person in any video for a reference image while keeping the scene and motion. Real models, real fixes, real output video.Fri, 03 Jul 2026 00:00:00 GMTsavienhttps://comfylab.dev/blog/workflows/wan-2-1-i2v-vs-ltxv-2-3-comfyui-same-scene-test/https://comfylab.dev/blog/workflows/wan-2-1-i2v-vs-ltxv-2-3-comfyui-same-scene-test/Took the exact first frame from a previous LTXV-2.3 text-to-video test and animated it with Wan 2.1 I2V instead -- a real, controlled comparison of two different local video generation approaches on the same RTX 3090.Fri, 03 Jul 2026 00:00:00 GMTsavienhttps://comfylab.dev/blog/workflows/wan-2-2-i2v-vs-ltxv-2-3-comfyui-same-scene-test/https://comfylab.dev/blog/workflows/wan-2-2-i2v-vs-ltxv-2-3-comfyui-same-scene-test/Ran the exact same image-to-video test with Wan 2.2's new dual-model MoE architecture instead of Wan 2.1 -- faster, and visibly sharper output. Real numbers from a local RTX 3090.Fri, 03 Jul 2026 00:00:00 GMTsavienhttps://comfylab.dev/blog/workflows/wan-22-video-generation-comfyui-image-to-video/https://comfylab.dev/blog/workflows/wan-22-video-generation-comfyui-image-to-video/Animate any image in ComfyUI with Wan 2.2. Which model size to pick for your GPU, node setup, and motion prompts that actually work.Fri, 03 Jul 2026 00:00:00 GMTsavien