Qwen Image: MiniMax Ref2VA FL2VA — 3 Workflows, Read From the Graphs
This release ships 3 workflow files built around Qwen Image. Together they total 126 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+Ref2VA|多参考全能视频生成 |
59 | 101 | euler |
— |
MiniMax+H3+FL2VA|文生+首尾帧全能视频 (1) |
39 | 51 | euler |
— |
MiniMax+H3|通用视频二阶高清重绘 |
28 | 45 | euler |
— |
The node count range is wide: MiniMax+H3|通用视频二阶高清重绘 at 28 nodes against MiniMax+H3+Ref2VA|多参考全能视频生成 at 59. 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_latent_upscaler_3d_fp16.safetensors |
MinimaxH3LatentUpscalerNode3D |
minimax_h3_fl2va_pruned_int8_convrot.safetensors |
UNETLoader |
minimax_h3_hybrid_fl2va_ref2va_b25-49-int8.safetensors |
UNETLoader |
minimax_h3_video_vae_int8_convrot.safetensors |
VAELoader |
LoRAs in play
| LoRA | Weight |
|---|---|
minimax_h3_fl2v_turbo_8step_v1.0_comfyui_bf16.safetensors |
1 |
minimax_h3_fl2v_turbo_4step_v1.0_768p_comfyui_bf16.safetensors |
1 |
Sampler choice
Samplers used: euler.
A plain euler is deterministic — good when the constraint is the point and noise per step would only perturb the trajectory. Ancestral variants inject noise each step and are the right pick when you want the model to explore rather than commit.
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
3 graphs published on RunningHub with invite code rh-v1111:
MiniMax+H3+FL2VA|文生+首尾帧全能视频 (1)MiniMax+H3+Ref2VA|多参考全能视频生成MiniMax+H3|通用视频二阶高清重绘
Files
https://huggingface.co/AI-KSK/ksk-video-296-comfyui-workflows
Video walkthrough: https://youtu.be/WKGrtR3p_0U
Bilibili: https://www.bilibili.com/video/BV1Bw8w6yEko/
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)