Qwen-Image WF — 4-8 Steps
this 8 Steps WorkFlow has 2 nodes that are absolutely necessary to squeeze the max quality of this kind of fast image generation: 1) Qwen-Image has somehow strange suggested official image sizes for generation:...
Runtime profile
- Access
- External workflow
- Status
- Listed
- Version
- Qwen-Image 8 Steps.json
- Source
- Civitai
- Updated
- Oct 14, 2026
Source description
this 8 Steps WorkFlow has 2 nodes that are absolutely necessary to squeeze the max quality of this kind of fast image generation:
- Qwen-Image has somehow strange suggested official image sizes for generation: "1:1": (1328, 1328), "16:9": (1664, 928), "9:16": (928, 1664), "4:3": (1472, 1140), "3:4": (1140, 1472), "3:2": (1584, 1056), "2:3": (1056, 1584) ... so for them it uses the "EmptyLatentImageCustom" node
- "ClownsharKSampler" node for proper access to additional samplers/schedulers + bongmath by RES4LYF
also you will need to get 4 or 8 steps lora: https://huggingface.co/lightx2v/Qwen-Image-Lightning/tree/main
What you need
- Model files
- 0.2 GiB known subtotal
Remaining hardware or download-size estimates are not available yet.
Requirements
Models
4qwen_image_vae.safetensors
Repository matchedVAE · Unknown
Qwen_Image-Q4_K_M.gguf
Not resolvedCheckpoint · Unknown
- Not resolved
LoRA · Unknown
Qwen2.5-VL-7B-Abliterated-Caption-it.Q5_K_M.gguf
Not resolvedText encoder · Unknown
Node packs
2comfyui-gguf
PossibleNode pack · Registry
RES4LYF
FoundNode pack · Registry
Sources
1 source
Sources
1 sourceModel artifacts
1 artifact
Model artifacts
1 artifactSource excerpts
1 excerpt
Source excerpts
1 excerptSource context: 6946 downloads · Type Workflows · Base model Qwen
