{"models":[{"endpoint_id":"minimax/h3-max/image-to-video","metadata":{"display_name":"H3 Max Image to Video","category":"image-to-video","description":"fal's H3 Max is a post-trained variant of MiniMax H3, tuned for stronger prompt adherence and better aesthetics while co-optimized with our custom inference stack for higher throughput with no compromises on output quality","status":"active","tags":["stylized","transform","lipsync"],"updated_at":"2026-10-24T02:47:44.558Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aa7658f/fSQlWMcduHn4_abgJ7qhQ.jpg","model_url":"https://fal.run/minimax/h3-max/image-to-video","license_type":"commercial","date":"2026-08-23T18:01:01.388Z","group":{"key":"hailuo-03-new-turbo","label":"Image to Video"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3-max-turbo/image-to-video","metadata":{"display_name":"H3 Max Turbo Image to Video","category":"image-to-video","description":"fal's H3 Max Turbo is a post-trained variant of MiniMax H3, tuned for stronger prompt adherence and better aesthetics while co-optimized with our custom inference stack for higher throughput with no compromises on output quality","status":"active","tags":["stylized","transform","lipsync"],"updated_at":"2026-10-18T22:17:00.576Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aa8d532/tsMg9lPG9DaXI7bqzqz6Y.jpg","model_url":"https://fal.run/minimax/h3-max-turbo/image-to-video","license_type":"commercial","date":"2026-09-02T16:51:51.105Z","group":{"key":"hailuo-03-new-max-turbo","label":"Image to Video Max Turbo"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3-max/reframe-video","metadata":{"display_name":"H3 Max Reframe","category":"video-to-video","description":"Expands existing videos to a selected aspect ratio with generative outpainting, preserving source audio and timing. Supports optional text guidance and 480p or 768p output.","status":"active","tags":["video","outpainting","reframing"],"updated_at":"2026-10-20T19:59:20.144Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aae1b2b/GYNF6LxWxgUK9WD5iYGBV.jpg","model_url":"https://fal.run/minimax/h3-max/reframe-video","license_type":"commercial","date":"2026-10-12T16:48:40.520Z","group":{"key":"hailuo-03-new-turbo","label":"Reframe Video"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3-max/reference-director","metadata":{"display_name":"H3 Max Reference Director","category":"worlds","description":"Direct a continuous world in real time with image references and text prompts.","status":"active","tags":["worlds","world-model","realtime","interactive","directable","reference-images","h3-max"],"updated_at":"2026-10-18T22:07:09.966Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aae847b/1Cd3rMhGJTSrHta846Gin.jpg","model_url":"https://fal.run/minimax/h3-max/reference-director","license_type":"commercial","date":"2026-10-15T20:25:05.488Z","group":{"key":"hailuo-03-new-turbo","label":"Reference Director"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3-max/reference-to-video","metadata":{"display_name":"H3 Max Reference to Video","category":"image-to-video","description":"fal's H3 Max is a post-trained variant of MiniMax H3, tuned for stronger prompt adherence and better aesthetics while co-optimized with our custom inference stack for higher throughput with no compromises on output quality","status":"active","tags":["stylized","transform","typography"],"updated_at":"2026-10-23T00:37:07.747Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aa84d7e/Weqd1RIL2EmDlzxnkQzCq.jpg","model_url":"https://fal.run/minimax/h3-max/reference-to-video","license_type":"commercial","date":"2026-08-29T16:21:32.325Z","group":{"key":"hailuo-03-new-turbo","label":"Reference to Video"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3-max/text-to-video","metadata":{"display_name":"MiniMax H3 Max Text to Video","category":"text-to-video","description":"fal's H3 Max is a post-trained variant of MiniMax H3, tuned for stronger prompt adherence and better aesthetics while co-optimized with our custom inference stack for higher throughput with no compromises on output quality","status":"active","tags":["stylized","transform","lipsync"],"updated_at":"2026-10-24T02:43:33.845Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aa7657e/LlavC7cCdZ_GGyvyglA82.jpg","model_url":"https://fal.run/minimax/h3-max/text-to-video","license_type":"commercial","date":"2026-08-23T17:56:01.222Z","group":{"key":"hailuo-03-new-turbo","label":"Text to Video"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3/image-to-video","metadata":{"display_name":"MiniMax H3 Image to Video","category":"image-to-video","description":"MiniMax H3 is a frontier video model. This endpoint animates a supplied image into 2K video, using it as the opening frame or pairs a first and last frame to control a transition between two images with the aspect ratio following the input.","status":"active","tags":["stylized","transform","lipsync"],"updated_at":"2026-08-25T17:41:07.462Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aa42dac/wjm-S2eWgG8vU2uJoRwPx.jpg","model_url":"https://fal.run/minimax/h3/image-to-video","license_type":"commercial","date":"2026-07-31T05:54:55.532Z","group":{"key":"hailuo-03-new","label":"Image to Video"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3/reference-to-video","metadata":{"display_name":"MiniMax H3 Reference to Video","category":"image-to-video","description":"MiniMax H3 is a frontier video model. This endpoint generates 2K video from multimodal references up to 9 images for subject and style, 3 video clips for motion, and 3 audio clips each cited in the prompt by order, keeping subjects consistent while following the referenced motion and audio.","status":"active","tags":["stylized","transform","lipsync"],"updated_at":"2026-08-25T17:41:07.462Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aa42dcf/3VDKabd7ifjCezrQpdfEI.jpg","model_url":"https://fal.run/minimax/h3/reference-to-video","license_type":"commercial","date":"2026-07-31T05:55:07.987Z","group":{"key":"hailuo-03-new","label":"Reference to Video"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3-max-turbo/text-to-video","metadata":{"display_name":"H3 Max Turbo Text to Video","category":"text-to-video","description":"fal's H3 Max Turbo is a post-trained variant of MiniMax H3, tuned for stronger prompt adherence and better aesthetics while co-optimized with our custom inference stack for higher throughput with no compromises on output quality","status":"active","tags":["stylized","transform","lipsync"],"updated_at":"2026-10-18T22:16:53.458Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aa8d50b/vj4BiCoQwOlmK7ku4KBZq.jpg","model_url":"https://fal.run/minimax/h3-max-turbo/text-to-video","license_type":"commercial","date":"2026-09-02T16:44:59.129Z","group":{"key":"hailuo-03-new-max-turbo","label":"Text to Video Max Turbo"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3-max/lip-sync/image-to-video","metadata":{"display_name":"H3 Max Lip Sync Image to Video","category":"image-to-video","description":"H3 Max Lip Sync generates a video from an image and supplied audio, synchronizing mouth movements to the soundtrack. It supports optional transcription guidance and output resolutions from 480p to 2K.","status":"active","tags":["lipsync","animation","audio"],"updated_at":"2026-10-20T22:39:43.329Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aaac3b8/AFRNL61Lm2CQntBtg88Hv.jpg","model_url":"https://fal.run/minimax/h3-max/lip-sync/image-to-video","license_type":"commercial","date":"2026-09-17T08:31:16.184Z","group":{"key":"hailuo-03-new-turbo","label":"Lip Sync"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3-max/recast","metadata":{"display_name":"H3 Max Recast","category":"video-to-video","description":"Recast the people in a video using reference photos with H3 Max, while preserving the source motion, camera, cuts, and audio.","status":"active","tags":["video-to-video","video-editing","character-replacement"],"updated_at":"2026-10-21T20:31:41.459Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aacabd7/P2ptkl7_WVxobibHV7Ti7.jpg","model_url":"https://fal.run/minimax/h3-max/recast","license_type":"commercial","date":"2026-10-01T18:51:18.250Z","group":{"key":"hailuo-03-new-turbo","label":"Recast"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3/text-to-video","metadata":{"display_name":"MiniMax H3 Text to Video","category":"text-to-video","description":"MiniMax H3 is a frontier video model. This endpoint generates video from a text prompt alone, rendering at 2K in durations from 5 to 15 seconds across seven aspect ratios.","status":"active","tags":["stylized","transform","lipsync"],"updated_at":"2026-08-26T06:41:51.586Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aa42d8c/uJC3CQO20PyPi5iM7eu3a.jpg","model_url":"https://fal.run/minimax/h3/text-to-video","license_type":"commercial","date":"2026-07-31T05:56:07.158Z","group":{"key":"hailuo-03-new","label":"Text to Video"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3-max/camera-controls","metadata":{"display_name":"H3 Max Camera Controls","category":"image-to-video","description":"H3 Max Multi Angle turns a single image into a video with precise, keyframe-based control over the camera's orbit, elevation, and distance in 3D space","status":"active","tags":["stylized","transform","editing"],"updated_at":"2026-10-18T22:16:09.827Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aaa0943/TDxACZJeJzny7u-Q-tV7w.jpg","model_url":"https://fal.run/minimax/h3-max/camera-controls","license_type":"commercial","date":"2026-09-11T15:56:44.375Z","group":{"key":"hailuo-03-new-turbo","label":"Camera Controls"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3-max/extend-video","metadata":{"display_name":"H3 Max Extend Video","category":"video-to-video","description":"H3 Max Extend Video adds a text-guided continuation to an existing video. It supports prompt expansion, adjustable duration and aspect ratio, and output resolutions from 480p to 2K, returning either the full extended video or only the new footage.","status":"active","tags":["extend","video","continuation"],"updated_at":"2026-10-19T14:46:27.347Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aaac3c2/Kw78XBxyzztvFFG4978eo.jpg","model_url":"https://fal.run/minimax/h3-max/extend-video","license_type":"commercial","date":"2026-09-17T08:33:13.639Z","group":{"key":"hailuo-03-new-turbo","label":"Extend Video"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3-max/3d-to-video","metadata":{"display_name":"H3 Max 3D to Video","category":"video-to-video","description":"Transform Blender renders and 3D previs into photorealistic video. H3 Max uses the source clip to guide scene layout, camera movement, and timing, with optional image references for appearance.","status":"active","tags":["3d","photorealistic","previs","video-to-video"],"updated_at":"2026-10-20T22:39:38.541Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aaad6ca/MkosWDWpYq9MCbeM0oopg_e4SyohXH.png","model_url":"https://fal.run/minimax/h3-max/3d-to-video","license_type":"commercial","date":"2026-09-16T16:39:24.294Z","group":{"key":"hailuo-03-new-turbo","label":"3D to Video"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3-max-turbo/extend-video","metadata":{"display_name":"H3 Max Turbo Extend Video","category":"video-to-video","description":"Extend an existing video with H3 Max Turbo: add 1 to 15 seconds of prompt-guided footage, with optional audio references and output resolutions from 480p to 2K.","status":"active","tags":["extend","video","continuation"],"updated_at":"2026-10-18T22:12:42.791Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aacaeb7/jUm3mOJpvKxTWhpLVoXCJ.jpg","model_url":"https://fal.run/minimax/h3-max-turbo/extend-video","license_type":"commercial","date":"2026-10-01T21:40:32.625Z","group":{"key":"hailuo-03-new-max-turbo","label":"Extend Video Max Turbo"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3-max/relight","metadata":{"display_name":"H3 Max Relight","category":"video-to-video","description":"Relight any video with H3 Max from a lighting sphere image, while preserving the source subjects, motion, camera and audio.","status":"active","tags":["video-to-video","video-editing","relighting"],"updated_at":"2026-10-19T14:48:16.105Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aad7408/RK0vzvak7-_A8Ajqd5pCC.jpg","model_url":"https://fal.run/minimax/h3-max/relight","license_type":"commercial","date":"2026-10-07T18:00:12.510Z","group":{"key":"hailuo-03-new-turbo","label":"Relight"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3/reference-to-video/lora","metadata":{"display_name":"MiniMax H3 Reference to Video LoRA","category":"video-to-video","description":"References into video with synchronized audio using MiniMax H3","status":"active","tags":["utility","editing"],"updated_at":"2026-08-25T17:43:52.103Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aa5d1a9/U9LZw7fmLzSjY6ycSnNYY.jpg","model_url":"https://fal.run/minimax/h3/reference-to-video/lora","license_type":"commercial","date":"2026-08-10T20:13:09.052Z","group":{"key":"hailuo-03-lora","label":"Reference to Video LoRA"},"highlighted":false,"kind":"inference","pinned":false,"training_endpoint_ids":["minimax/h3/flf2v/trainer","minimax/h3/i2v/trainer"]}},{"endpoint_id":"minimax/h3/image-to-video/lora","metadata":{"display_name":"MiniMax H3 Image to Video LoRA","category":"image-to-video","description":"Images into video with synchronized audio using MiniMax H3; your image becomes the first frame, prompt optional, with trained LoRA support for subject consistency.","status":"active","tags":["utility","editing"],"updated_at":"2026-08-25T17:43:52.103Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aa4c825/EewvNmxzIkXlC4vNqsPgi.jpg","model_url":"https://fal.run/minimax/h3/image-to-video/lora","license_type":"commercial","date":"2026-08-02T23:27:10.009Z","group":{"key":"hailuo-03-lora","label":"Image to Video LoRA"},"highlighted":false,"kind":"inference","pinned":false,"training_endpoint_ids":["minimax/h3/flf2v/trainer","minimax/h3/i2v/trainer"]}},{"endpoint_id":"minimax/h3-max/insert-video","metadata":{"display_name":"H3 Max","category":"video-to-video","description":"Insert a new scene into an existing video with H3 Max. Guide the scene with a prompt, reference images, or reference videos, then return to the original footage.","status":"active","tags":["video-to-video","video-editing","scene-insertion"],"updated_at":"2026-10-19T14:46:56.018Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aac45ed/CUlIKgQ55O4HiSkSG0OEL.jpg","model_url":"https://fal.run/minimax/h3-max/insert-video","license_type":"commercial","date":"2026-09-28T23:44:04.870Z","group":{"key":"hailuo-03-new-turbo","label":"Insert Video"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3-max/styles/hand-drawn","metadata":{"display_name":"H3 Max Hand Drawn","category":"text-to-video","description":"Generates 768p video with audio in a hand-drawn animation style from text prompts or an optional first-frame image. Supports durations of 5–15 seconds.","status":"active","tags":["hand-drawn","pencil","animation","stylized"],"updated_at":"2026-10-18T22:14:33.961Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aaad5f2/3kPTumERGySlFQrvJdc-5.jpg","model_url":"https://fal.run/minimax/h3-max/styles/hand-drawn","license_type":"commercial","date":"2026-09-17T21:29:26.890Z","group":{"key":"hailuo-03-new-turbo-lora","label":"Hand Drawn"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3-max/styles/vhs","metadata":{"display_name":"H3 Max VHS","category":"text-to-video","description":"Generates 768p VHS-style video with audio from text prompts or an optional first-frame image. Supports 5–15 second clips and adjustable tape damage, from subtle analog noise to strong tracking distortion.","status":"active","tags":["vhs","analog","retro","tape-damage"],"updated_at":"2026-10-18T22:14:02.074Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aaad605/J4zSNIgmNAoArcg4BwPNp.jpg","model_url":"https://fal.run/minimax/h3-max/styles/vhs","license_type":"commercial","date":"2026-09-17T21:32:21.470Z","group":{"key":"hailuo-03-new-turbo-lora","label":"VHS"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3/text-to-video/lora","metadata":{"display_name":"MiniMax H3 Text to Video LoRA","category":"text-to-video","description":"Generate video with synchronized audio from a text prompt using MiniMax H3; load a trained LoRA at adjustable strength to lock in style, character, or motion.","status":"active","tags":["utility","editing"],"updated_at":"2026-09-01T21:35:15.577Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aa4c826/eJSEtWE-CowTbabIAcO5f.jpg","model_url":"https://fal.run/minimax/h3/text-to-video/lora","license_type":"commercial","date":"2026-08-02T23:26:20.888Z","group":{"key":"hailuo-03-lora","label":"Text to Video LoRA"},"highlighted":false,"kind":"inference","pinned":false,"training_endpoint_ids":["minimax/h3/t2v/trainer"]}},{"endpoint_id":"minimax/h3-max/styles/16bit-pixel","metadata":{"display_name":"H3 Max 16-bit Pixel","category":"text-to-video","description":"Generates 768p video with audio in a 16-bit pixel-art style from text prompts or an optional first-frame image. Supports durations of 5–15 seconds.","status":"active","tags":["pixel-art","16-bit","animation","stylized"],"updated_at":"2026-10-18T22:14:37.674Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aaad5e9/M4QpbIMC5EcjkbBo4AW33.jpg","model_url":"https://fal.run/minimax/h3-max/styles/16bit-pixel","license_type":"commercial","date":"2026-09-17T21:28:00.908Z","group":{"key":"hailuo-03-new-turbo-lora","label":"16-bit Pixel"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3/ref2va/trainer","metadata":{"display_name":"MiniMax H3 Reference to Video LoRA Trainer","category":"training","description":"Train a MiniMax H3 LoRA with reference conditioning, so different modalities animate into video with audio; captions optional.","status":"active","tags":[],"updated_at":"2026-09-01T21:34:14.751Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aa5ce43/QHlY-BQzMKU1t0r1e5cGs.jpg","model_url":"https://fal.run/minimax/h3/ref2va/trainer","license_type":"commercial","date":"2026-08-10T17:48:29.973Z","group":{"key":"minimax-h3-trainer","label":"Reference to Video"},"highlighted":false,"kind":"training","pinned":false}},{"endpoint_id":"minimax/h3-max/styles/retro-toon-70s","metadata":{"display_name":"H3 Max Retro Toon 70s","category":"text-to-video","description":"Generates 768p video with audio in a retro 1970s hand-painted animation style from text prompts or an optional first-frame image. Supports durations of 5–15 seconds.","status":"active","tags":["retro","1970s","hand-painted","animation"],"updated_at":"2026-10-18T22:14:03.956Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aaad5fe/vIqC2SSPOPaR305av-iHV.jpg","model_url":"https://fal.run/minimax/h3-max/styles/retro-toon-70s","license_type":"commercial","date":"2026-09-17T21:31:46.115Z","group":{"key":"hailuo-03-new-turbo-lora","label":"Retro Toon 70s"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3-max/styles/low-poly","metadata":{"display_name":"H3 Max Low Poly","category":"text-to-video","description":"Generates 768p video with audio in a retro low-poly 3D style from text prompts or an optional first-frame image. Supports durations of 5–15 seconds.","status":"active","tags":["low-poly","retro","3d-animation","stylized"],"updated_at":"2026-10-18T22:14:29.660Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aaad5f8/7YH1svyhRf42Itvh-gQ4T.jpg","model_url":"https://fal.run/minimax/h3-max/styles/low-poly","license_type":"commercial","date":"2026-09-17T21:30:25.689Z","group":{"key":"hailuo-03-new-turbo-lora","label":"Low Poly"},"highlighted":false,"kind":"inference","pinned":false}},{"endpoint_id":"minimax/h3/i2v/trainer","metadata":{"display_name":"MiniMax H3 Image to Video Audio LoRA Trainer","category":"training","description":"Train a MiniMax H3 LoRA with first-frame conditioning, so a still image animates into video with audio; captions optional.","status":"active","tags":["utility","editing"],"updated_at":"2026-10-18T16:20:13.876Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aa4c7e2/g_bTE07BPKDYnHM51ySM-.jpg","model_url":"https://fal.run/minimax/h3/i2v/trainer","license_type":"commercial","date":"2026-08-02T23:13:09.898Z","group":{"key":"minimax-h3-trainer","label":"Image to Video-Audio"},"highlighted":false,"kind":"training","pinned":false,"inference_endpoint_ids":["minimax/h3/image-to-video/lora"]}},{"endpoint_id":"minimax/h3/t2v/trainer","metadata":{"display_name":"MiniMax H3 Text to Video LoRA Trainer","category":"training","description":"Train a MiniMax H3 LoRA on your own captioned clips for pure text-to-video generation with matching audio.","status":"active","tags":["utility","editing"],"updated_at":"2026-09-01T21:34:14.751Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aa4c7ed/p7BcKn3HYb-pOF2U4XZjX.jpg","model_url":"https://fal.run/minimax/h3/t2v/trainer","license_type":"commercial","date":"2026-08-02T23:14:40.505Z","group":{"key":"minimax-h3-trainer","label":"Text to Video-Audio"},"highlighted":false,"kind":"training","pinned":false,"inference_endpoint_ids":["minimax/h3/text-to-video/lora"]}},{"endpoint_id":"minimax/h3/flf2v/trainer","metadata":{"display_name":"MiniMax H3 First Last Frame LoRA Trainer","category":"training","description":"Train a MiniMax H3 LoRA on first/last/both keyframe signatures, teaching it to generate video with audio that starts on one image and lands on another.","status":"active","tags":["utility","editing"],"updated_at":"2026-09-01T21:34:14.751Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aa4c7e5/uSvYTt7kgW68f5wSHkQco.jpg","model_url":"https://fal.run/minimax/h3/flf2v/trainer","license_type":"commercial","date":"2026-08-02T23:13:31.815Z","group":{"key":"minimax-h3-trainer","label":"First&Last Frame to Video-Audio"},"highlighted":false,"kind":"training","pinned":false,"inference_endpoint_ids":["minimax/h3/image-to-video/lora"]}},{"endpoint_id":"minimax/h3-max/director","metadata":{"display_name":"H3 Max Director","category":"worlds","description":"Direct continuous, realtime video streams with live prompts while preserving characters, settings, and story continuity.","status":"active","tags":[],"updated_at":"2026-10-18T22:16:38.616Z","is_favorited":null,"thumbnail_url":"https://v3b.fal.media/files/b/0aa8e1ed/B0XPwpNjLEqrbapCIGPJj.jpg","model_url":"https://fal.run/minimax/h3-max/director","license_type":"commercial","date":"2026-09-03T17:02:49.768Z","group":{"key":"hailuo-03-new-turbo","label":"Director"},"highlighted":false,"kind":"inference","pinned":false}}],"next_cursor":null,"has_more":false}