Filmmaker turns two low-res refs into cinematic frames with Minimax-Image in 5 minutes
GrungeWerX · reddit · 2026-11-18
A filmmaking-focused Reddit user shares a hands-on test of Minimax-Image, calling it the first model that takes color-grading direction well and produces a genuine film aesthetic.
Key points:
- The cinematic frame was generated from just two low-quality inputs: one anime character sheet and a tiny low-res background thumbnail, with character likeness preserved
- No LoRAs were used — only a public 20MP workflow (JSON shared on Hugging Face) with the prompt enhancer stripped out, plus LTX-2.5 for video; the whole process took 5 minutes
- Prompts stayed short, centered on "Create a real photo of the character in Picture 1," with minor details on features, stage, grade and "cinematic 50mm Sony" framing
- The author plans to use LTX-2.5 for b-roll and test Minimax video shots next
More from Multimodal
- Suno launches Visuals: AI music now gets auto-lipsynced videos and prompt-made artwork — The Verge AI · 2026-11-18
- AI-built 3D racing game lands on official OpenAI blog — Dimillian · 2026-11-18
- AI-generated retro CRT computer image is nearly indistinguishable — LinusEkenstam · 2026-11-18
- Wes Roth builds a 4-minute AI animated film end-to-end in Genspark Media Studio — Wes Roth · 2026-11-18
- Former voice actor launches Voyager 1, an AI voice you can direct line by line — Scobleizer · 2026-11-18
- Recreating Whiplash's tense rehearsal room with MiniMax H3 and ComfyUI — hucun16 · 2026-11-18