MiniMax H3 ComfyUI workflow generates consistent character sheets, beats Qwen on quality
cgpixel23 · reddit · 2026-10-04
Reddit user cgpixel23 shared a ComfyUI workflow that uses MiniMax H3 as an image generation model to build detailed, multi-view character sheets from a single reference image or a batch, meant to improve character consistency in downstream video generation.
Usage is simple: load the MiniMax H3 models and reference images, pick the step count (8 steps for speed, 25 for best quality), and run. Benchmarks against Qwen Image 2.1:
- Quality: H3 wins, outputting 4096x1536 vs Qwen's 1664x928, with slightly better character consistency
- Speed: Qwen is faster (6 steps, 4.5 min vs H3 Turbo Fused 25 steps, 12 min)
The workflow (Google Drive + Civitai article) and a video tutorial are publicly available.
More from Multimodal
- Why AI filmmakers keep making creepy visuals and calling them great films — taherdhanera · 2026-10-05
- One-shot agentic video editing: Opus 5.5 turns a footage folder and brief into a finished UGC ad — alexgoughcooper · 2026-10-05
- Blogger predicts cinemas won't survive next decade as AI lets people make movies at home — taherdhanera · 2026-10-05
- AI video: Back to the Future DeLorean flies through 100 years of history — mark_k · 2026-10-05
- Open workflow extends static video shots without degradation using Ref2v, masks and bridge resampling — roychodraws · 2026-10-05
- AI Short Puts Two Michael Jackson Personas on One Mic for a Rap Duet — SimplyAnnisa · 2026-10-04