MiniMax R2V Local Test: 27 Minutes to Generate a 5-Second Video
cocktailpeanut · x · 2026-08-06
A developer successfully ran MiniMax's R2V (image-to-video) model locally in ComfyUI via Pinokio. Testing showed it took about 27 minutes to generate a short 5-second video clip. The author noted that prompting is crucial for success and plans to explore ways to optimize inference time.
More from Multimodal
- MiniMax Video Model's True Strength Lies in Long-Form Storytelling — cocktailpeanut · 2026-08-06
- Animating Static Images with MiniMax H3: Local Workflow & Parameters — y3kdhmbdb2ch2fc6vpm2 · 2026-08-06
- MiniMax H3 in Practice: Native Audio-Video Generation and Advanced Prompting — Smyshnikof · 2026-08-06
- H3 Model Generates Blocky and Distorted Images, Users Debug Node Settings — FastIce8391 · 2026-08-06
- Minimax video model runs locally: 4080 generates 1280x720 in ~30 mins — cocktailpeanut · 2026-08-06
- FLUX3's edge: 20 seconds generation time — cocktailpeanut · 2026-08-06