Fashion video made locally with MiniMax H3: h3.c cuts Apple Silicon gen time 2-3x
MindfulPornographer · reddit · 2026-09-24
A creator shares a high-fashion video generated locally on an M5 Max: first frame from Cyberrealistic Z-Image 8.0 via Draw Things, motion from MiniMax H3. They find H3's prompt adherence and motion clearly better than LTX, but stock ComfyUI generation is too slow — using h3.c, optimized for Apple Silicon, cuts generation time by 2-3x.
More from Multimodal
- Generating accurate 3D building models from 150 planning PDFs with new Gemini model — fofrAI · 2026-11-11
- Stanford's DyCraft Uses Agentic Decomposition to Cut Physical Errors in Long-Horizon 4D Scene Generation — StanfordUniversity · 2026-11-11
- Midjourney SREF pick: --2523103199 for loose pencil-sketch portraits — tisch_eins · 2026-11-11
- H3 Image Tips: 16MP via ComfyUI Workflow Plus Detailed Prompts Yields Best Results — gatortux · 2026-11-11
- Gemini-generated endless loop of quantum cryostats self-assembling — hapticdata · 2026-11-11
- Dad brings his daughter's crayon drawings to life with AI for $1.75, and it's unsettling — Illustrious_Ad_4292 · 2026-11-11