About ComfyLab

What is ComfyLab?

ComfyLab is a technical blog dedicated to ComfyUI and AI content generation: images, video and audio. Our goal is for anyone, regardless of technical level, to follow a tutorial and get real results on the first try.

The author

My name is Savien. I've spent years working with generative AI tools, and I originally built ComfyLab because there wasn't a quality Spanish-language resource on ComfyUI — this English edition extends the same hands-on, verified approach to a wider audience.

My main rig is an RTX 3090 (24GB VRAM), though every tutorial states the minimum GPU needed to reproduce it (many work fine with 4-8GB VRAM).

My experience

• Image generation with Stable Diffusion and ComfyUI since 2022
• Extensive testing with SDXL, Flux, and video pipelines like AnimateDiff and Wan 2.1
• Optimizing workflows for low-VRAM GPUs (4-8GB)
• Tracking the ComfyUI ecosystem since its early versions

Content philosophy

Every article includes:

• Exact, reproducible steps
• The correct values for every node or parameter
• The minimum required GPU
• Reference screenshots or videos where needed

We don't publish vague guides. If a tutorial says "set CFG to 7," there's a reason, and we explain it.

Contact

Questions, suggestions, or want to collaborate? Reach out through the comments on any article.