MiniMax H3, a video model, makes surprisingly good 4–16 MP stills in 6 steps
ReferenceFluffy6159 · reddit · 2026-11-17
A free ComfyUI workflow (H3 Image AIO) shows MiniMax H3, a video model, doubling as a text-to-image and editing tool, with backgrounds showing real depth instead of dreamy blur.
Key takeaways from testing:
- The editing LoRA gives the biggest quality jump and fewer bad draws; slightly Eastern look in non-editing mode
- A 6-step sigma schedule works where the official numbers failed—keep the workflow's values
- SeedVR2 at 1× only adds detail, since H3 renders 4–16 MP natively
- Prompt expansion is optional and consumes most of the runtime; disable it for finished prompts
The workflow supports up to 6 reference images/video, a reserved trigger-word box, color adjustment, and 4 example prompts. Author's score: Qwen Image 2.1 remains the daily driver (9/10); H3 Image 8.5/10 with text rendering close to MOSS Image.
More from Multimodal
- Seedance 2.5 4K video generation goes live on fal, US-hosted inference — gorkem · 2026-11-17
- Seedance 2.5 demo shows Hollywood-level cinematic action realism — SimplyAnnisa · 2026-11-17
- RelayVSR open-source 4x video upscaler: 144p clip upscaled in ~2 min on a 4070 — Ant_6431 · 2026-11-17
- Suno pulls off noise/math-rock drums and atonal piano on request — AndyMasley · 2026-11-17
- Seedance2.5-generated Ramayana manga chapter: Jatayu's last stand — CurieuxExplorer · 2026-11-17
- Creator adapts the entire Ramayana into a manga using Higgsfield's Mangaka — CurieuxExplorer · 2026-11-17