The official Z-Image Turbo (bf16) with the Comfy-Org template settings: res_multistep / simple, 8 steps, CFG 1, shift 3, 1024x1024. Distilled for few steps, so keep CFG at 1 and leave the negative empty.The official Z-Image Turbo bf16 model with the settings of the Comfy-Org Z-Image Turbo template.Every wire runs through collapsed Set/Get pairs, green titles mark what you are meant to edit, one Seed node drives every stage, prompts are saved by Prompt Manager and LoRAs go in the LoRA Manager loader. Optional groups (bypassed by default) are toggled in Toggle stages: Download models and RTX Upscale (2x RTX Video Super Resolution, ULTRA 0.5; needs ComfyUI_Nvidia_VFX_Nodes and an NVIDIA RTX GPU, so leave it off otherwise).Files: <code>kiko-zimage-turbo-v6.json</code><h2 id="custom-nodes">Custom Nodes</h2><ul><li><a target="_blank" rel="ugc" href="https://github.com/ComfyAssets/ComfyUI-KikoTools">ComfyUI-KikoTools</a></li><li><a target="_blank" rel="ugc" href="https://github.com/kijai/ComfyUI-KJNodes">ComfyUI-KJNodes</a></li><li><a target="_blank" rel="ugc" href="https://github.com/willmiao/ComfyUI-Lora-Manager">ComfyUI-Lora-Manager</a></li><li><a target="_blank" rel="ugc" href="https://github.com/ComfyAssets/ComfyUI_PromptManager">ComfyUI_PromptManager</a></li><li><a target="_blank" rel="ugc" href="https://github.com/rgthree/rgthree-comfy">rgthree-comfy</a></li></ul>Optional:<ul><li><a target="_blank" rel="ugc" href="https://github.com/Comfy-Org/ComfyUI_Nvidia_VFX_Nodes">ComfyUI_Nvidia_VFX_Nodes</a>: only for the bypassed RTX Upscale group; needs an NVIDIA RTX GPU. Without it, ComfyUI warns about a missing node on load; keep the group bypassed</li></ul><h2 id="models">Models</h2><ul><li><code>z_image_turbo_bf16.safetensors</code> → <code>models/diffusion_models/</code>: <a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/z_image_turbo/resolve/main/split_files/diffusion_models/z_image_turbo_bf16.safetensors">https://huggingface.co/Comfy-Org/z_image_turbo/resolve/main/split_files/diffusion_models/z_image_turbo_bf16.safetensors</a></li><li><code>qwen_3_4b.safetensors</code> → <code>models/text_encoders/ZImage/</code>: <a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/z_image_turbo/resolve/main/split_files/text_encoders/qwen_3_4b.safetensors">https://huggingface.co/Comfy-Org/z_image_turbo/resolve/main/split_files/text_encoders/qwen_3_4b.safetensors</a></li><li><code>z-image-vae.safetensors</code> → <code>models/vae/ZImage/</code>: <a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/Z-Image-Turbo/resolve/main/split_files/vae/ae.safetensors">https://huggingface.co/Comfy-Org/Z-Image-Turbo/resolve/main/split_files/vae/ae.safetensors</a></li></ul><h2 id="first-run">First run</h2>Each workflow has a bypassed Download models group. Turn it on in Toggle stages, queue once, then turn it off again. Files you already have are skipped. Civitai models and gated Hugging Face models need a token, set where ComfyUI starts (never in the workflow):<pre><code># Linux / macOS (~/.bashrc, ~/.zshrc or your start script)
export CIVITAI_API_KEY="your-civitai-key"
export HF_TOKEN="hf_your-token"

# Windows (Command Prompt once, then a new window)
setx CIVITAI_API_KEY "your-civitai-key"
setx HF_TOKEN "hf_your-token"</code></pre>Restart ComfyUI after setting them.<h2 id="example-and-workflow-image">Example and workflow image</h2>The example image was rendered with <code>kiko-zimage-turbo-v6.json</code>, all optional groups at their defaults (Prompt Enhancer, RTX Upscale and Download models off). Prompt and seed are in <code>prompt.txt</code>.<code>kiko-zimage-turbo-workflow.png</code> (in the download) is a picture of the whole graph, exported with ComfyUI-Custom-Scripts' Workflow Image → Export → png. It has the workflow embedded: drag it into ComfyUI to load it. Prompt boxes and image previews are page elements, so they show empty in the picture.Happy rendering! Kiko

Files with SHA256: cfbcdb32d47a3e150fcf33db3481b02a611666933a8999ed22b9d596f1a6563b