First anime image-to-video test with MiniMax H3 in ComfyUI shows promise, artifacts remain
Kuttachuuu · reddit · 2026-09-05
A creator shared a first real attempt at anime-style image-to-video using MiniMax H3 via ComfyUI, with starting frames generated by Krea 2 and the focus on H3's motion quality.
Self-assessment: motion is not yet perfect, and micro-movements still produce artifacts. The author is iterating and soliciting advanced prompts and tips specifically for MiniMax H3.
More from Multimodal
- AI video now pulls off orbiting, diving camera moves around a speeding bicycle — umesh_ai · 2026-10-23
- AI art showcase: gothic winged warrior "Azrael-42" with porcelain-crack armor — LudovicCreator · 2026-10-23
- Moody AI-reimagined Harry Potter portraits showcase stunning cinematic generation — azed_ai · 2026-10-23
- Claude Code + ComfyUI MCP workflow for client work: paid models or open weights? — newaccount47 · 2026-10-23
- AI-generated character sheet wows: melancholy traveling cat 'Finch' fully spec'd — azed_ai · 2026-10-23
- Viral prompt template turns AI image models into children's book character-sheet generators — azed_ai · 2026-10-23