{"path":"/compare/kitten-dance-motion-transfer","title":"Higgsfield Genjutsu vs Seedance 2.0 Mini vs MiniMax H3: can a kitten copy a man’s dance? | Fourth Feed","description":"We asked three video models to put a small kitten into a man’s car alarm dance. Genjutsu copied the moves but made the kitten human-sized; Seedance and MiniMax kept it small but lost the moves.","as_of":"2026-11-08","kind":"compare","content":{"compare":{"slug":"kitten-dance-motion-transfer","title":"Higgsfield Genjutsu vs Seedance 2.0 Mini vs MiniMax H3: can a kitten copy a man’s dance?","description":"We asked three video models to put a small kitten into a man’s car alarm dance. Genjutsu copied the moves but made the kitten human-sized; Seedance and MiniMax kept it small but lost the moves.","query":"higgsfield genjutsu","asOf":"2026-11-08","intro":"Motion transfer models take a reference video and put someone else into it. We wanted a small kitten, a 3-month-old about 40 to 55 cm tall on its hind legs, to do the viral car alarm dance that TV meteorologist Nick Kosir posted on TikTok in January 2024: a grown man in a grey suit hopping, kicking and swinging his arms next to a white SUV. On 2026-11-08 we gave the same dance and the same kitten photo to three models. The short answer: none of them did both things at once. Higgsfield Genjutsu (standard motion transfer, 480p) copied the dance almost move for move, but drew the kitten as tall as the man, with its helmet above the SUV’s tail lights. Writing the size into the prompt (40 to 55 cm, shorter than the tyre) and giving it a first frame with the kitten at the right size changed nothing. Genjutsu’s API has no size setting, and the open-source copy Openjutsu works by turning the original dancer into a mannequin of the same outline, which would explain why the new character fills a human-sized shape. Seedance 2.0 Mini (reference-to-video, 720p, the full 12 seconds) kept the kitten small in every frame when we passed the small first frame as a reference image. The man never appeared and the camera and timing followed the original, but the kitten swapped his kicks and arm swings for generic bouncing and ran on all fours where he crouched. Describing every move second by second in the prompt made it stiffer, not closer. MiniMax H3 (multimodal reference, 768P, 5 seconds; it bills the reference video’s seconds as well as the output’s) also kept the kitten small, and also lost the moves: where the man hops on one leg with an arm over his head, the kitten walks toward the camera with its paws out. Our reading: these models read the moves from the reference dancer’s body. When the new character is shaped like the old one, the moves carry over; when a kitten has to copy a grown man, a model either keeps his body size or keeps only his rhythm. Getting the first frame right was easy by comparison: an image model needed three tries, and the size only came out right once we anchored it to the car (the top of the helmet at the wheel hub), not in centimetres. Through the APIs, each test clip cost between $0.58 and $0.71.","video":{"src":"/compare/kitten-dance-motion-transfer-five-panels.mp4","poster":"/compare/kitten-dance-motion-transfer-five-panels.jpg","captions":"/compare/kitten-dance-motion-transfer-five-panels.vtt","caption":"The first 4 seconds of the reference dance (Nick Kosir, TikTok, 2024-01-25) and the same seconds from each test, with the original sound. Genjutsu’s clip came back 3.7 seconds long and is stretched to match. “Mini + pose prompt” is the same Seedance 2.0 Mini run with every move written into the prompt."},"ai":[],"ours":{"name":"Pugsoda","url":"https://pugsoda.com/templates/car-alarm-dance-ai","text":"If you just want your cat or dog in the car alarm dance: our Car Alarm Dance template takes one photo and gives pets, small dogs and kids a small dancer to copy, so they stay their real size."},"author":"eric-lo","sponsored":false,"tools":["higgsfield.ai","byteplus.com","minimax.io","hailuoai.video"],"tested":{"higgsfield.ai":{"result":"Genjutsu standard motion transfer at 480p from a 4-second reference. Copied the dance closely, but the kitten came out as tall as the man. A size in the prompt and a correctly sized first frame did not change that. Output ran about 7% shorter than the reference.","checkedAt":"2026-11-08"},"byteplus.com":{"result":"Seedance 2.0 Mini reference-to-video at 720p, 12 seconds. With the small first frame as the first reference image, the kitten stayed small and the man never appeared, but its moves were generic bouncing rather than his.","checkedAt":"2026-11-08"},"minimax.io":{"result":"MiniMax H3 multimodal reference (images plus a reference video) at 768P on the official API: billed 10 seconds for a 5-second clip (5 in, 5 out). Kept the kitten small, lost the moves, and pushed the camera in faster than the original.","checkedAt":"2026-11-08"}}},"author":{"id":"eric-lo","name":"Eric Lo","bio":"Eric works on Pugsoda, a site of video templates that put you or your pet into a trending clip from one photo.","product":{"name":"Pugsoda","url":"https://pugsoda.com"}},"rows":[{"tool":{"domain":"higgsfield.ai","name":"Higgsfield","url":"https://higgsfield.ai/","summary":"Creative suite for generating and editing images, video and voice from prompts or references.","category":["text-to-video"],"tags":["image-to-video","ai-video-editor"],"logo":{"file":"/logos/higgsfield.ai.webp","fetchedAt":"2026-11-08"},"source":["research:sitedata-2026-10-18"],"site":{"title":"Higgsfield AI — AI-native creative suite","description":"Create images, videos, and voice content with Higgsfield AI from text prompts or references. Edit and upscale media, automate creative workflows with its AI agent, and generate content on web and mobile.","llmsTxt":true,"checkedAt":"2026-10-29","alive":true},"agentReady":{"llmsTxt":true,"mcpRegistry":null},"traffic":{"monthlyVisits":40561218,"month":"2026-09","source":"sitedata.dev traffic ranking, 2026-10-18"},"domainCreated":"2023-06-07","firstSeen":"2026-10-29","referral":{"url":"https://higgsfield.ai?fpr=eric-85ed25","youGet":"Up to 50% off a plan, valid for 3 hours after you first log in through the link.","weGet":"Up to 25% of the plans you buy, including renewals and upgrades, for up to 12 months, paid in cash.","terms":"https://higgsfield.ai/affiliate","checkedAt":"2026-11-10"}},"categoryNames":["Text and image to video"],"hasPage":true},{"tool":{"domain":"byteplus.com","name":"BytePlus","url":"https://byteplus.com/","summary":"ByteDance’s enterprise cloud, which sells API access to its own models such as Seedance.","category":["video-api"],"tags":["video-model-api","seedance-api"],"logo":{"file":"/logos/byteplus.com.webp","fetchedAt":"2026-11-08"},"source":["research:compare-seedance-2-api-2026-10-29"],"site":{"title":"BytePlus | AI-Native Cloud for Enterprise Growth","description":"Build, deploy, and scale AI with BytePlus — cutting-edge models, tools, and infrastructure proven on global platforms.","llmsTxt":true,"checkedAt":"2026-10-29","alive":true},"pricing":{"model":"paid","freeTier":"Trial credits for verified business accounts","from":"$0.07 per second of Seedance 2.0 video at 480p (provider page, 2026-10-23)"},"agentReady":{"llmsTxt":true,"mcpRegistry":null},"firstSeen":"2026-10-29","notes":"Strengths: official first-party cloud endpoint for ByteDance Seedance and Seaweed video models. Trade-offs: enterprise onboarding and region-specific console setup compared with unified pay-per-call gateways."},"categoryNames":["Video model APIs and data APIs"],"hasPage":true},{"tool":{"domain":"minimax.io","name":"MiniMax","url":"https://minimax.io/","summary":"Model lab behind the Hailuo video and MiniMax language models, with consumer apps and an API.","category":["video-api"],"tags":["video-model-api"],"logo":{"file":"/logos/minimax.io.webp","fetchedAt":"2026-11-08"},"source":["research:sitedata-2026-10-18"],"site":{"title":"MiniMax","description":"Building AGI with our mission Intelligence with Everyone. Global leader in multi-modal models and AI-native products with over 300 million users.","llmsTxt":false,"checkedAt":"2026-10-29","alive":true},"pricing":{"model":"freemium"},"agentReady":{"llmsTxt":false,"mcpRegistry":null},"traffic":{"monthlyVisits":6050028,"month":"2026-09","source":"sitedata.dev traffic ranking, 2026-10-18"},"domainCreated":"2021-03-27","firstSeen":"2026-10-29"},"categoryNames":["Video model APIs and data APIs"],"hasPage":true},{"tool":{"domain":"hailuoai.video","name":"Hailuo AI","url":"https://hailuoai.video/","summary":"Generates videos and images from text or photos with MiniMax Hailuo models.","category":["text-to-video"],"tags":["image-to-video"],"logo":{"file":"/logos/hailuoai.video.webp","fetchedAt":"2026-11-08"},"source":["research:sitedata-2026-10-18"],"site":{"title":"Hailuo AI: MiniMax H3 LIVE NOW. Top-Tier Quality, Versatile References","description":"Create AI videos and images from text or photos with Hailuo AI. Explore video, image, and meme tools built for creators and social content online.","llmsTxt":true,"checkedAt":"2026-10-29","alive":true},"agentReady":{"llmsTxt":true,"mcpRegistry":null},"traffic":{"monthlyVisits":2268830,"month":"2026-09","source":"sitedata.dev traffic ranking, 2026-10-18"},"domainCreated":"2024-09-05","firstSeen":"2026-10-29"},"categoryNames":["Text and image to video"],"hasPage":true}]}}