Reference Videos Make Minimax Acting Believable: Full Prompt Template Shared
roychodraws · reddit · 2026-10-10
A Reddit user shows how reference videos dramatically improve acting quality in Minimax video generation: characters copy the delivery reference's cadence, pauses, pitch changes, emphasis, tone and facial manner.
The key trick is separating appearance from performance: each subject's looks come exclusively from a designated picture, while delivery style comes from a reference video. The template explicitly instructs the model to ignore the reference speaker's appearance, voice timbre, words, setting and watermarks, and includes a structured 'delivery style: fullypreserved' block to lock in the performance transfer.
Related event: Reddit Users Discover Workflows for Natural Acting in MiniMax-H3(3 posts)→
More from Multimodal
- Alibaba's TaoMate-H3 streams minute-long video with audio using 3-step denoising — jiqizhixin · 2026-10-10
- SGLang-Diffusion serving framework for diffusion models to be unveiled at PyTorch Conference 2026 — PyTorch · 2026-10-10
- image-blaster repo turns a single photo into a fully explorable 3D world in five minutes — anthara_ai · 2026-10-10
- MoMA commissions Holly Herndon and Mat Dryhurst for dual-venue AI art show — matdryhurst · 2026-10-10
- Open-source creative harness Voyager adds 8 languages, drives Blender, Resolve and 100+ tools — ravisparikh · 2026-10-10
- Full AI video made at home on a single RTX 3090 with Qwen, FLUX, MiniMax and YuE2 — solyarisoftware · 2026-10-10