Loading
Have a ComfyUI workflow? Get a readable Workflow Doc in your browser.Make a workflow readable.Try it now ↗
This workflow allows users to generate a coarse image using Chroma and then refine it with the Z-Image-Turbo model.
Runtime profile
Source description
This workflow allows users to generate a coarse image using Chroma and then refine it with the Z-Image-Turbo model.
Regarding Denoise Strength: If you are generating NSFW images or subjects that Z-Image might not recognize, you should keep the denoise around 0.2-0.3 to preserve details. Alternatively, use 0.6 or higher to make the images more realistic.
Tip: If you want to reduce VRAM usage, simply use the same checkpoint for both model loaders.
This is just my personal workflow, intended only as a rough reference. It's not perfect, but I hope it serves as a starting point to inspire better ideas. Huge thanks to Lodestone Lodestone Creator Profile | Civitai for providing such an amazing model.
AI-generated commentary
AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.
Use this workflow to create a refined image: it generates a coarse image with Chroma, passes that image to Z-Image-Turbo, and outputs the refinement.
The image sent to refinement is the coarse image generated with Chroma; the output is that image after Z-Image-Turbo refinement.
The workflow lists these model files: ae.safetensors, Chroma-DC-2K-BF16.gguf, chroma/lenovo.safetensors, Chroma1-Base.safetensors, Chroma1-HD-fp8_scaled_rev2.safetensors, and flux-dev-vae.safetensors.
Other listed model files are qwen_3_4b.safetensors, t5xxl_fp16.safetensors, t5xxl_fp8_e4m3fn_scaled.safetensors, z_image_turbo_bf16.safetensors, and zimg_vae.safetensors.
The named node packs are ComfyLiterals, ComfyUI-GGUF, and RES4LYF.
For NSFW images or subjects Z-Image may not recognize, the creator suggests denoise strength around 0.2–0.3 to preserve details; for a more realistic look, the creator suggests 0.6 or higher.
Suggestion · not verified
To reduce VRAM use, the creator suggests using the same checkpoint in both model loaders.
Suggestion · not verified
The workflow names ComfyUI-GGUF; confirm the intended package before setup.
Suggestion
The source states no use license for this workflow; check the applicable terms before using it.
Suggestion
Does this need to be edited?
Sign in to send an edit request.
Sources
1 sourceSource excerpts
1 excerptSource context: 289 downloads · Type Workflows · Base model Chroma
Remaining hardware or download-size estimates are not available yet.
Requirements
VAE · Hugging Face · Comfy-Org/Lumina_Image_2.0_Repackaged · Lumina Image 2.0 Repackaged
Chroma-DC-2K-BF16.gguf
Not resolvedCheckpoint · Unknown
LoRA · 45.1 MB · SAFETENSORS · Hugging Face · chroma/lenovo.safetensors
Checkpoint · 16.6 GB · SAFETENSORS · Unknown
Chroma1-HD-fp8_scaled_rev2.safetensors
File unverifiedCheckpoint · Hugging Face · silveroxides/Chroma1-HD-fp8-scaled · Chroma1 HD FP8 Scaled
flux-dev-vae.safetensors
Not resolvedVAE · Unknown
Text encoder · 7.49 GB · SAFETENSORS · Unknown
Text encoder · 9.12 GB · SAFETENSORS · Unknown
t5xxl_fp8_e4m3fn_scaled.safetensors
File unverifiedText encoder · Hugging Face · comfyanonymous/flux_text_encoders · Flux Text Encoders
Checkpoint · 11.5 GB · SAFETENSORS · bf16 · Unknown
zimg_vae.safetensors
Not resolvedVAE · Unknown
ComfyLiterals
FoundNode pack · Registry
ComfyUI-GGUF
PossibleNode pack · Registry
RES4LYF
FoundNode pack · Registry
