minimax/h3-max-turbo/text-to-video

{
  "video": {
    "url": "https://v3b.fal.media/files/b/0aa8d548/1oeWwPEh_I_QHYRxp70af_minimax-h3.mp4",
    "content_type": "video/mp4",
    "file_name": "1oeWwPEh_I_QHYRxp70af_minimax-h3.mp4",
    "file_size": 3919321
  },
  "expanded_prompt": "integrated_multimodal_description: [Shot 1] Cinematic 2D motion design with a Swiss editorial aesthetic. The scene opens on a solid, warm ivory background. Large, bold black typography reading \"H3 MAX TURBO\" slides in rapidly from the left edge of the frame, decelerating with a clean ease-out to center. Simultaneously, a thin vermilion red line grows horizontally from the center outward underneath the text. As the text settles, the entire composition compresses slightly along the Y-axis and snaps back instantly with a spring-like motion. [Shot 2] At 00:02.500, the camera cuts to a dynamic typographic sequence. The red line extends aggressively across the entire width of the screen, acting as a sliding crop transition that reveals massive, stacked black text: \"FASTER\" on the top line and \"THAN REAL TIME.\" on the bottom. The typography is oversized, with edges cropped by the frame boundaries. The text moves with fast horizontal shifts and sharp landings, maintaining a high-contrast palette of ivory, black, and vermilion red. [Shot 3] At 00:05.500, the typography collapses through a fast mask transition into five rectangular video-frame cards arranged in a shallow 2.5D perspective. The cards fill with content rapidly one after another in a stacking sequence, while a red timing line underneath moves at a significantly slower pace. The camera executes a whip-pan movement with clean parallax, visually emphasizing that the generation process is finishing ahead of the timing line. [Shot 4] At 00:08.500, the five cards snap together with a sudden motion-design impact into one flat, 2D composition. A fast mask reveal displays the text: \"5s VIDEO.\" and \"GENERATED IN 1.5s.\". The \"1.5s\" is rendered as the largest visual element on screen, appearing with a strong scale-up pop. [Shot 5] At 00:11.000, a vermilion red line shoots around the perimeter of the frame, building a rectangular border. The previous text is pushed out of the frame by the moving crop of the border. The border then contracts and expands several times in a rhythmic, pulsing motion. [Shot 6] At 00:12.700, the screen returns to a static warm ivory background. After a momentary pause, a final heavy impact reveal displays the text \"now live on fal.\" in clean, lowercase black typography, centered within a thin red rectangular frame. The composition is completely flat and still, holding without movement until the end.\n\noverall_soundscape: A sequence of high-impact foley sounds including a sharp \"whoosh\" for the sliding text, a metallic \"snap\" for the compression, and a heavy, dampened \"thud\" when the final text lands. A rhythmic, accelerating heartbeat-like bass pulse accompanies the contracting border. The final reveal is preceded by a sudden, absolute silence.\n\nnon_diegetic_music: A minimal, high-energy electronic track characterized by a driving, syncopated bassline and sharp percussive hits. The tempo accelerates during the card-stacking sequence and cuts abruptly to silence at 12.7 seconds before one final, resonant bass hit on the closing frame.",
  "timings": {
    "inference": 1.4260040810331702
  }
}