Re-render a source image with a Z-Image checkpoint at a chosen denoise strength: the 6B base (CFG and negative prompt) or a Civitai Turbo finetune on its 8-step schedule, with an optional Z-Image LoRA. No upstream Z-Image image-to-image template exists at Comfy-Org/workflow_templates@aaac56dd, so the graph is this catalog's chroma-img2img shape (LoadImage, VAEEncode, KSampler at a denoise below 1) with the Z-Image model stack and sampler values of templates/image_z_image.json (res_multistep + simple, 25 steps, CFG 4, ModelSamplingAuraFlow shift 3) and templates/image_z_image_turbo.json (8 steps, CFG 1, negative zeroed); the source is bounded to a longest side of 1024 with the ImageScaleToMaxDimension node templates/image_z_image_turbo_fun_union_controlnet.json uses. Choosing a Turbo checkpoint selects the Turbo schedule through checkpoint_profile.
Tags: zimagerunpod-serverlessvolume-backedimage-to-image
Inputs (12)#
The typed parameter surface callers bind when they request this workflow. Enum options and numeric bounds are the values the workflow document declares.
promptstringrequireddefault a close-up photograph of the same scene at golden hour, warm light, natural colours, sharp focusnegative_promptstringdefault checkpointenumdefault zimage-base-bf16zimage-base-bf16cyberrealistic-zimage-turbo-v8-bf16zimage-turbo-bf16-aiostepsintegerdefault 25min 8max 60cfgfloatdefault 4.0min 1.0max 10.0seedintegerdefault -1source_imageimagerequireddenoisefloatdefault 0.6min 0.05max 1.0loraenumdefault nonenonephotoreal-betternudes-v3realistic-snapshot-v5breast-sliderdetail-sliderdetaildaemonz-v1.1lora_strengthfloatdefault 1.0min -2.0max 2.0detailenumdefault nonenonefacehandsface_and_handsdetail_denoisefloatdefault 0.4min 0.05max 1ComfyUI node graph (15)#
The executable ComfyUI prompt graph: 15 nodes across 14 distinct node classes, wired by 8 data dependencies. Nodes tinted green come from a custom node pack this workflow declares; the rest are ComfyUI core / baked-community classes.
Nodes (15)#
1UNETLoadercoreunet_name = {{checkpoint_map[checkpoint]}} tmplweight_dtype = defaultMODEL2CLIPLoadercoreclip_name = qwen_3_4b.safetensorstype = lumina2device = defaultCLIP3VAELoadercorevae_name = ae.safetensorsVAE4CheckpointLoaderSimplecoreckpt_name = {{checkpoint_map[checkpoint]}} tmplMODELCLIPVAE5LoraLoaderModelOnlycoremodel = {{loader_refs[checkpoint_profile[checkpoint].loader].model}} tmpllora_name = {{lora_map[lora]}} tmplstrength_model = {{lora_strength}} tmplMODEL6ModelSamplingAuraFlowcoremodel = {{lora_ref if lora != 'none' else loader_refs[checkpoint_profile[checkpoint].loader].model}} tmplshift = 3MODEL7CLIPTextEncodecoretext = {{constructed_prompt}} tmplclip = {{loader_refs[checkpoint_profile[checkpoint].loader].clip}} tmplCONDITIONING8CLIPTextEncodecoretext = {{negative_prompt}} tmplclip = {{loader_refs[checkpoint_profile[checkpoint].loader].clip}} tmplCONDITIONING9ConditioningZeroOutcoreconditioning = ◂ node 7 · out[0]CONDITIONING10LoadImagecoreimage = {{source_image}} tmplIMAGEMASK11ImageScaleToMaxDimensioncoreimage = ◂ node 10 · out[0]upscale_method = lanczoslargest_size = 1024IMAGE12VAEEncodecorepixels = ◂ node 11 · out[0]vae = {{loader_refs[checkpoint_profile[checkpoint].loader].vae}} tmplLATENT13KSamplercoremodel = ◂ node 6 · out[0]positive = ◂ node 7 · out[0]negative = {{checkpoint_profile[checkpoint].negative}} tmpllatent_image = ◂ node 12 · out[0]seed = {{seed}} tmplsteps = {{distilled_schedule.steps if checkpoint_profile[checkpoint].distilled else steps}} tmplcfg = {{distilled_schedule.cfg if checkpoint_profile[checkpoint].distilled else cfg}} tmplsampler_name = res_multistepscheduler = simpledenoise = {{denoise}} tmplLATENT14VAEDecodecoresamples = ◂ node 13 · out[0]vae = {{loader_refs[checkpoint_profile[checkpoint].loader].vae}} tmplIMAGE15SaveImagecoreimages = ◂ node 14 · out[0]filename_prefix = isis/zimage-img2imgPrompt construction#
template{base_prompt}Variables (1)#
base_prompt{{prompt}} tmplParameter banks (9)#
The prompt / configuration lookup tables this workflow keys into from its inputs — the vocabulary that turns a style / palette / preset selection into graph parameters.
option_policy (2)#
checkpoint{"zimage-base-bf16": {"distribution": "service", "content": "neutral"}, "cyberrealistic-zimage-turbo-v8-bf16": {"distribution": "service", "content": "racy"}, "zimage-turbo-bf16-aio": {"distribution": "service", "content": "neutral"}}lora{"photoreal-betternudes-v3": {"distribution": "service", "content": "explicit"}, "realistic-snapshot-v5": {"distribution": "service", "content": "neutral"}, "breast-slider": {"distribution": "service", "content": "racy"}, "detail-slider": {"distribution": "service", "content": "neutral"}, "detaildaemonz-v1.1": {"distribution": "service", "content": "neutral"}}checkpoint_map (3)#
zimage-base-bf16z_image_bf16.safetensorscyberrealistic-zimage-turbo-v8-bf16cyberrealistic_zimage_turbo_v8_bf16.safetensorszimage-turbo-bf16-aiozimage_turbo_bf16_aio.safetensorscheckpoint_profile (3)#
zimage-base-bf16{"loader": "unet", "distilled": false, "negative": ["8", 0]}cyberrealistic-zimage-turbo-v8-bf16{"loader": "unet", "distilled": true, "negative": ["9", 0]}zimage-turbo-bf16-aio{"loader": "aio", "distilled": true, "negative": ["9", 0]}distilled_schedule (2)#
steps8cfg1.0loader_refs (2)#
unet{"model": ["1", 0], "clip": ["2", 0], "vae": ["3", 0]}aio{"model": ["4", 0], "clip": ["4", 1], "vae": ["4", 2]}lora_ref (2)#
50lora_map (6)#
nonephotoreal-betternudes-v3zimage-photoreal-betternudes-v3.safetensorsrealistic-snapshot-v5zimage-realistic-snapshot-v5.safetensorsbreast-sliderzimage-breast-slider.safetensorsdetail-sliderzimage-detail-slider.safetensorsdetaildaemonz-v1.1zimage-detaildaemonz-v1.1.safetensorspost_render (1)#
detail_tail{"detail_input": "detail", "denoise_input": "detail_denoise", "decode_node": "14", "frame_from_image": {"input": "source_image", "max_dimension": 1024}, "seed": {"node": "13", "input": "seed"}, "model": {"node": "13", "input": "model"}, "positive": {"node": "13", "input": "positive"}, "negative": {"node": "13", "input": "negative"}, "steps": {"node": "13", "input": "steps"}, "cfg": {"node": "13", "input": "cfg"}, "sampler_name": {"node": "13", "input": "sampler_name"}, "scheduler": {"node": "13", "input": "scheduler"}, "working_scale": 1.5, "size_multiple": 16, "max_working_pixels": 4194304, "max_faces": 4, "max_hands": 8, "mask_grow": 16, "mask_blur": 4, "sam3_checkpoint": "sam3.1_multiplex_fp16.safetensors", "sam3_threshold": 0.5, "mask_prefix": "isis/zimage-img2img-detail-mask"}requires_families (4)#
zimagezimage-cyberrealisticzimage-aiosam31Models & dependencies#
Models required (11)#
z_image_bf16.safetensorscyberrealistic_zimage_turbo_v8_bf16.safetensorszimage_turbo_bf16_aio.safetensorsqwen_3_4b.safetensorsae.safetensorszimage-photoreal-betternudes-v3.safetensorszimage-realistic-snapshot-v5.safetensorszimage-breast-slider.safetensorszimage-detail-slider.safetensorszimage-detaildaemonz-v1.1.safetensorssam3.1_multiplex_fp16.safetensorsOutput contract#
What a successful run of this workflow returns.
typeimageformatpngTaxonomy & routing#
How the control plane classifies this workflow — from the committed workflow-taxonomy-registry.json. It drives the consistency / control surface the agentic director can exercise over the workflow.
assetFamilyportrait-hero-image-bundleoutputPackageProfileimage-single-profilecontrolModalitiesmodel-locksampler-scheduler-lockseed-locklatent-reuseconsistencyDimensionsidentitylightingenvironmentnotesZ-Image image-to-image at a chosen denoise; a Turbo checkpoint selects the 8-step schedule.