Qwen Image: MiniMax ControlNet Union — 5 Workflows, Read From the Graphs
This release ships 5 workflow files built around Qwen Image. Together they total 457 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 |
|---|---|---|---|---|
MiniMax+H3|ControlNet+Union++局部重绘专用工作流 |
184 | 306 | — |
— |
MiniMax H3|ControlNet Union 2.0 视频控制与局部重绘|VDN |
120 | 189 | — |
— |
MiniMax+H3|ControlNet+Union+2.0+视频控制与局部重绘·音频驱动 |
66 | 99 | — |
— |
MiniMax H3|通用视频生成·音频驱动数字人|VDN 12步高清版 |
44 | 80 | — |
— |
MiniMax H3|通用视频生成|VDN 12步高清版 |
43 | 73 | — |
— |
The node count range is wide: MiniMax H3|通用视频生成|VDN 12步高清版 at 43 nodes against MiniMax+H3|ControlNet+Union++局部重绘专用工作流 at 184. 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 |
|---|---|
qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors |
CLIPLoader |
minimax_h3_video_vae_fp16.safetensors |
VAELoader |
minimax_h3_audio_vae_fp32.safetensors |
VAELoader |
minimax_h3_fl2va_pruned_int8_convrot.safetensors |
UNETLoader |
minimax_h3_latent_upscaler_3d_fp16.safetensors |
MiniMaxH3LearnedLatentUpscaleT8Advanced |
minimax_h3_fun_controlnet_union_pruned_int8_convrot.safetensors |
ModelPatchLoader |
minimax_h3_fl_ref_hybrid_scheme_a_25-49_bf16.safetensors |
UNETLoader |
minimax_h3_hybrid_fl2va_ref2va_b25-49-int8.safetensors |
UNETLoader |
minimax_h3_fun_controlnet_union_2.0_pruned_int8_convrot.safetensors |
ModelPatchLoader |
SeC-4B-bf16.safetensors |
SeCModelLoader |
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:
MiniMax H3|ControlNet Union 2.0 视频控制与局部重绘|VDN 12步高清版MiniMax H3|通用视频生成·音频驱动数字人|VDN 12步高清版MiniMax H3|通用视频生成|VDN 12步高清版MiniMax+H3|ControlNet+Union++局部重绘专用工作流MiniMax+H3|ControlNet+Union+2.0+视频控制与局部重绘·音频驱动数字人|VDN+12步高清版
Files
https://huggingface.co/AI-KSK/ksk-video-315-comfyui-workflows
Video walkthrough: https://youtu.be/WdAaPq2Xa0I
Bilibili: https://www.bilibili.com/video/BV1SEJm6PE22/
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)