Loop ForgeSanitize: Jin's page
From MOSS-Image add-ons, tested locally
- Model
- MOSS-Image Base (moss_image_base_int8_convrot.safetensors)
- Output
- 864×1152
- Sampling
- 30 steps, CFG 4, er_sde_cns / normal
- Seed
- 20
- Without LoRA
- 104.0 s, peak VRAM 10.1 GiB
- Sanitize
- 60.0 s, peak VRAM 10.7 GiB
- Ran on
- RTX 3060, 12GB (local)
What was wired in
Input page
Jin's comic page (sanitize)Comic page drawn from the Jin cast photo with the Qwen-Image 2.1 edit template (not MOSS)Download PNG1.4 MB
Across the three pages it took the speech bubbles out every time, and it cleared a faint watermark spread across one of the pictures. The big sound effects and the copyright line in the corner stayed, so check what it leaves behind before you use it.
no sfx, no dialogue, no watermark
Notes
Sanitize is its own LoRA (moss_i2i_sanitize_v3) on MOSS Base, loaded by the MOSS i2i Reference node together with the input page. The prompt is the LoRA's fixed caption, as its helper writes it.
Times on this page are for these two runs.
sorryhyun/MOSS-I2I-LORAs on Hugging Face
Run it yourself
Download everything for this test (.zip, 1.4 MB). It holds the workflow, every input it loads and a README. Copy the inputs into ComfyUI/input, put the models below in place and drag the workflow onto ComfyUI.
| File | Put it in | Size |
|---|---|---|
| moss_image_base_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_i2i_sanitize_v3.safetensors | models/loras | 148.7 MB |
Custom nodes: ComfyUI-MOSS-Toolkit at commit 53ee3b3. Everything else is part of ComfyUI 0.43.1 (frontend 1.57.6, templates 0.11.100).