H3 LongTakeNoCuts: 3-minute no-cut video workflow with 19 stitched clips
xyzdist · reddit · 2026-10-15
A creator shared a full ComfyUI workflow for generating a 3-minute cinematic one-take video with H3: one character reference + ref2va, hybrid model b20-49, kitchen attention, no LoRA, 10 steps, 19x10s clips at 0.7mp (11 min each), stitched in CapCut. Key fixes: character masking to refine only the face and prevent drift, reseeding/locking seeds when cuts occur, the fizgig H3 tweak node (contrast -0.1, per-frame) to soften faces, and extracting frames as references to stabilize backgrounds on camera pull-backs.
More from Multimodal
- AI-Rendered Gotham After Dark: Retro Fabric-Print Bat Signals and Walls of Joker Faces — egeberkina · 2026-10-16
- Turn charts, logos and photos into animated videos in 5 minutes with Claude Code + Remotion — 6 copy-paste prompts — TawohAwa · 2026-10-16
- Kling 4.0's Omni Reference supports up to 15 reference assets in one generation — SarahAnnabels · 2026-10-16
- Kling 4.0 holds character and environment coherent in large-scale two-person climbing shots — SarahAnnabels · 2026-10-16
- Kling 4.0 keeps character stable across a massive desert scene with endless movement — SarahAnnabels · 2026-10-16
- Seedance 2.5 workflow: character reference sheets + structured prompt yields a 30s single-take fake tourist video — techhalla · 2026-10-16