FLUX: MOSS Image DMD Turbo — 6 Workflows, Read From the Graphs
This release ships 6 workflow files built around FLUX. Together they total 105 nodes. Rather than summarise the video, this write-up reads the graphs directly — what each one loads, how big its latents are, and which sampler it runs.
The workflows at a glance
| Workflow | Nodes | Links | Sampler | Latent |
|---|---|---|---|---|
洗图最终 (4) |
35 | 43 | — |
— |
MOSS Image|50步原版 vs 8步DMD vs 4步Turbo·三档加速横评版 |
20 | 27 | — |
1024x1024 1f |
MOSS Image|Base vs SFT 双模型同参对决·画质横评版 |
16 | 18 | — |
1024x1024 1f |
MOSS Image|色彩深度LoRA·光影与色彩控制实验版 |
12 | 11 | — |
1024x1024 1f |
MOSS Image|JSON Layout 结构化布局·图文排版控制版 |
11 | 10 | — |
1024x1024 1f |
MOSS Image|SFT 50步高质量文生图·新手入门版 |
11 | 10 | — |
1024x1024 1f |
The node count range is wide: MOSS Image|JSON Layout 结构化布局·图文排版控制版 at 11 nodes against 洗图最终 (4) at 35. That gap is usually the difference between a single-pass preview graph and one that adds conditioning, upscale stages, or a second sampling pass.
What the graphs load
| Model file | Loader node |
|---|---|
moss_image_sft_int8_convrot.safetensors |
UNETLoader |
flux2-vae.safetensors |
VAELoader |
qwen3.5_9b_bf16.safetensors |
CLIPLoader |
moss_image_base_int8_convrot.safetensors |
UNETLoader |
flux3_image_klein_sd_int8_convrot.safetensors |
UNETLoader |
qwen3vl_4b_int8_convrot.safetensors |
CLIPLoader |
RealESRGAN_x4plus.pth |
UpscaleModelLoader |
LoRAs in play
| LoRA | Weight |
|---|---|
moss_image_sft_4step_v2.0.safetensors |
1 |
MOSS_DMD_8STEP_QKV_FORGE_FIX.safetensors |
1 |
moss_image_color_depth_control.safetensors |
0.8 |
Not every LoRA runs at 1.0 — the non-default weights here are [0.8]. A weight above 1.0 amplifies the LoRA's effect past what its author trained for; below 1.0 backs it off. Both are deliberate tuning choices, not leftovers.
Latent geometry
Distinct latent configurations across the set: 1024x1024 1f.
The frame count matters more than the pixel dimensions here. A graph at 97 frames is built for roughly a four-second unit; one at 121 frames is closer to five. Sigma schedules do not transfer cleanly between them, so treat each geometry as its own tuning target.
Where this set is weak
- Node count is not quality. The largest graph is not automatically the best one; a 40-node graph with a good prompt beats a 200-node graph with a bad one.
- No measured comparison in the files. The workflows carry settings, not benchmarks. Any quality claim needs your own A/B run.
- Model files are not included. You supply the weights; the graphs only reference filenames.
- Version drift. A graph pinned to a specific checkpoint revision will run on a newer one but may produce different output.
Practical order
Open the smallest graph first. It has the fewest moving parts and will confirm your models load before you spend time on the elaborate ones.
Workflows
5 graphs published on RunningHub with invite code rh-v1111:
MOSS Image|50步原版 vs 8步DMD vs 4步Turbo·三档加速横评版MOSS Image|Base vs SFT 双模型同参对决·画质横评版MOSS Image|JSON Layout 结构化布局·图文排版控制版MOSS Image|SFT 50步高质量文生图·新手入门版MOSS Image|色彩深度LoRA·光影与色彩控制实验版洗图最终 (4)
Files
https://huggingface.co/AI-KSK/ksk-video-325-comfyui-workflows
Video walkthrough: https://youtu.be/09qEF5oo8Gc
Bilibili: https://www.bilibili.com/video/BV1ZHHP6FEVZ/
Notes
- Workflow files only. Model weights are not included and each keeps its own licence.
- Node counts, geometries and loader filenames above are read from the published JSON files.
- Independent AI-KSK release. Not affiliated with the model authors or RunningHub.
Optional API platform - 88API (affiliate link)
Explore 88API for access to multiple AI model APIs. Available models, routes and prices are listed on the platform; Alipay top-ups are supported.
https://88api.ai/sign-up?aff=usx3
Disclosure: I may earn a commission from eligible purchases after you register through this referral link.
Top comments (0)