MOSS Image SFT int8 workflow for clean, natural photos, with 2 free LoRA slots ready for your own LoRAs (off by default, Ctrl+B to enable).
Skip Layer Guidance on layer 22 for cleaner structure
Optional HD redraw with Ultimate SD Upscale 2x
Resolution picker: just choose a ratio + megapixels
Same clean design as my other workflows: numbered groups, group toggles, readable node titles
Built and tested on an RTX 4070 Laptop 8GB / 16GB RAM.
These workflows are built and improved with the help of Claude (Anthropic). The collection keeps evolving with it, so expect regular updates.
Name : 1.0 · MOSS × USD
--- Version changes or notes ---
Flow: generate (30 steps) -> save
-> USD 2x (group ②) -> color match -> save HD
Toggle each group on/off with the bars above the groups.
Settings:
Sampler: euler / sgm_uniform, 30 steps, CFG 3
Skip Layer Guidance: layer 22, from 0 to 50 %
Negative prompt: empty
Default size: 3:4 portrait, 1.0 MP (change it in “Resolution”)
USD: 2x with 4x Remacri, 20 steps, denoise 0.15, color match mkl 0.7
About the HD pass:
MOSS shines at its native 1024 px, so USD upscales straight from pass 1 for the cleanest result. A latent second pass is included but muted; press Ctrl+M on those nodes if you want to experiment with it.
Models:
mossImage_sft_pruned_int8.safetensors (UNET)
mossImage_sft_txt_int4.safetensors (text encoder, type moss_image)
flux2-vae.safetensors (VAE): https://huggingface.co/black-forest-labs/flux-3-image-klein-comfy/resolve/main/vae/flux2-vae.safetensors
4x_foolhardy_Remacri.pth (upscaler)
Custom nodes:
rgthree-comfy: https://github.com/rgthree/rgthree-comfy
KJNodes: https://github.com/kijai/ComfyUI-KJNodes
ComfyUI_UltimateSDUpscale: https://github.com/ssitu/ComfyUI_UltimateSDUpscale
Works with any recent ComfyUI with MOSS Image support.
These workflows are built and improved with the help of Claude (Anthropic): testing recipes, cleaning the files, writing the notes. The collection keeps evolving with it, so expect regular updates.Description
MOSS Yogi x USD
A MOSS workflow built around Realism by Stable Yogi, a photoreal MOSS fine-tune made for phone-camera candids, selfies, portraits and real light. No speed LoRA needed, the model is fast on its own.
How it works
Pass 1 generates the image at around 1.5 MP (2:3) in 10 steps, then a light sharpen finishes it. This is the main output, one image per run.
Two optional upgrades are off by default:
- Pass 2 (2K): upscales the image in pixel space with 4x Remacri, resizes it to x1.5 (around 2K), encodes it back and refines it with a light 0.35 denoise, which keeps it sharp instead of the blurry edges you get from a latent upscale. Switch it on with the "Second pass" toggle.
- USD: Ultimate SD Upscale 2x for a large final image. Switch it on with the "HD redraw" toggle.
Settings
10 steps, CFG 1, er_sde, beta (the settings recommended by Stable Yogi).
Models needed
Realism by Stable Yogi V1.5 (INT8-ConvRot, or the file that fits your card), mossImage_sft_txt_int4 (text encoder, type moss_image), flux2-vae (VAE), 4x_foolhardy_Remacri (upscaler).
Custom nodes: rgthree-comfy, KJNodes, ComfyUI_UltimateSDUpscale.
LoRA
Two LoRA nodes, bypassed by default. Press Ctrl+B on one and pick your own LoRA.