No LoRAs, no Upscalers, just prompting, executed in the simplest way possible thanks to <a target="_blank" rel="ugc" href="https://civarchive.com/models/2168935">"Z-Image Turbo"</a> and <a target="_blank" rel="ugc" href="https://github.com/martin-rizzo/ComfyUI-ZImagePowerNodes">"The Power Nodes"</a>. The result?... TOTAL FUN !!! 🎉🤩<h2 id="but-what-are-these-workflows">But What Are These Workflows?</h2>These are experimental workflows where I test visual styles and prompting techniques for Z-Image Turbo. Each workflow groups visual styles around a specific subject (e.g., Cinema, Sketch, Anime), so within each workflow, you will find 10+ preconfigured styles related to the main theme.<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/a3c11f74-01d7-4d1a-8581-e1f3e15420b4/original=true/a3c11f74-01d7-4d1a-8581-e1f3e15420b4.jpeg" />The workflows are specifically designed to leverage Nodes 2.0 and Sub-graphs within ComfyUI, while being organized to make image generation highly intuitive and user-friendly.<h2 id="requirements">Requirements</h2><ul><li><a target="_blank" rel="ugc" href="https://github.com/Comfy-Org/ComfyUI">ComfyUI:</a> v0.2.7 or higher</li><li><a target="_blank" rel="ugc" href="https://github.com/martin-rizzo/ComfyUI-ZImagePowerNodes">Z-Image Power Nodes:</a> v2.1.1 or higher</li></ul><h2 id="checkpoints-used">Checkpoints Used</h2><ul><li>ZIT: <a target="_blank" rel="ugc" href="https://huggingface.co/martin-rizzo/Z-Image-Turbo-INT8-ConvRot-ComfyUI/blob/main/z_image_turbo_int8_convrot_bf16emixed.safetensors">z_image_turbo_int8_convrot_bf16emixed</a></li><li>TE: <a target="_blank" rel="ugc" href="https://huggingface.co/martin-rizzo/Qwen3-4B-INT8-ConvRot-ComfyUI/blob/main/qwen3-4b_int8_convrot_fp16emixed.safetensors">qwen3-4b_int8_convrot_fp16emixed</a></li><li>VAE: <a target="_blank" rel="ugc" href="https://huggingface.co/easygoing0114/Z-Image_clear_vae/blob/main/Z-Image_half_natural_vae.safetensors">Z-Image_half_natural_vae</a></li></ul>Note: While these specific checkpoints are used in each workflow, any Z-Image Turbo checkpoint should work well, including the original models from the ComfyUI repository.

Z-Image Total Fun!!! 🎉🤩

v1: First release of the workflow dedicated to generating images that resemble stop-motion films, miniatures, and dioramas.

Files with SHA256: 3bf8d0c2248fb5efe38757c3a1e67a8514d281e36df2c26504688273dc98e217