Loop ForgeDMD 8 steps: the cafe sign
From MOSS-Image add-ons, tested locally
- Model
- MOSS-Image SFT (moss_image_sft_int8_convrot.safetensors)
- Output
- 1024×1024
- Seed
- 101
- Template, 25 steps
- 25 steps, CFG 2.5, er_sde / simple. 41.9 s, peak VRAM 10.4 GiB
- DMD LoRA, 8 steps
- 8 steps, CFG 1, er_sde / simple. 8.7 s, peak VRAM 10.3 GiB
- Ran on
- RTX 3060, 12GB (local)
Verdict
Almost five times faster. For drafts I'd use this every time; for a final image I'd still go back to the full steps.
Photograph of a man with short black hair, a trimmed beard and a white linen shirt with rolled sleeves, standing in front of a small corner cafe at golden hour. Above the door a hand-painted wooden sign reads "THE SLOW KETTLE" and a chalkboard beside him reads "Fresh bread at 7. Open late on Fridays." Warm light, shallow depth of field, 35mm photo.
nsfw, worst quality, bad anatomy, watermark, jpeg artifact
Notes
The official template with lodestones' DMD LoRA (lora.safetensors) at 1.0, 8 steps and CFG 1, next to the template's own 25 steps and CFG 2.5.
lodestones/MOSS-Image-SFT-8-steps-dmd on Hugging Face
Comfy-Org workflow templates (the MOSS-Image templates ship with ComfyUI)
Run it yourself
This test loads no input files. Put the models below in place and drag the workflow onto ComfyUI.
| File | Put it in | Size |
|---|---|---|
| moss_image_sft_int8_convrot.safetensors | models/diffusion_models | 5.4 GB |
| qwen3.5_9b_w4a8_gptq.safetensors | models/text_encoders | 7.3 GB |
| flux2-vae_small.safetensors | models/vae | 249.5 MB |
| moss_sft_8step_dmd.safetensors (published as lora.safetensors, renamed) | models/loras | 1.2 GB |
No custom nodes. Every node is part of ComfyUI 0.43.1 (frontend 1.57.6, templates 0.11.100).