Model
.ai
tracking open ai models

BUNNY General Motion Continuity Repair

8663°
BUNNY General Motion Continuity Repair
Created by
FourBunny
on Civitai
by FourBunny - This is a general-purpose motion support LoRA, not a combat-only LoRA. It can be used for running, sports , dance, acrobatics, character interaction, combat, weapon motion, and other dynamic scenes.
Creator notes
General Motion Continuity Repair LoRA 🎬 BUNNY Motion Repair V2 — High-Difficulty Motion Shouldn't Be a Lottery. This is a general-purpose motion support LoRA, not a combat-only LoRA. It can be used for running / sports / dance / acrobatics / character interaction / combat / weapon motion and other dynamic scenes. 这是一个通用动态辅助 LoRA,并不是战斗专用模型,可用于跑步 / 体育 / 舞蹈 / 翻滚跳跃 / 人物互动 / 战斗 / 武器动作等动态场景。 🚀 Run It on RunningHub 🎁 New users on RunningHub International can get 1000 RH Coins after registration. 国内用户可以使用: 🔥 What’s New in V2 - Targeting Failure Zones: Focuses on MiniMax H3's hardest movement areas—rapid flips, spins, rolls, inversions, and extreme orientation recovery. - Rotational & Inverted Upgrade: Significantly expands full-body rotational training while keeping V1’s general motion repair power. - Smooth Full Sequences: Enhances the complete chain: Takeoff ➔ Inversion ➔ Continuous Spins ➔ Reorientation ➔ Landing, not just the middle frames. 🤸‍♂️ Noticeable Motion Gains - Power Moves & Ground Spins: Thomas flares, windmills, airflares, and ground rotations stay stable without collapsing during repeated cycles. - Airborne Continuity: Flips, twisting jumps, and aerial spins maintain structural integrity—no more twisted body masses, misconnected limbs, or out-of-sync head-torso rotation. 💡 Under the Hood - Capability-Oriented Training: Learns structural stability, orientation awareness, and momentum recovery rather than memorizing trick names, ensuring strong generalization to unseen motions. - Usable Output First: As a repair LoRA, rare imperfect frames can still happen in extreme transitions, but it reliably shifts generations from discarded clips to usable shots. 💬 In Short High difficulty, higher stability. Flips stay coherent, spins stay together, and landings stop falling apart. (高难动作,稳定性跟上。空翻不扭、旋转不散、落地不崩!) Recommended Usage / 推荐使用 Standalone use: around 0.9 is currently recommended. 单独使用时,目前推荐权重约为 0.9。 When used alone, Motion Continuity Repair needs a relatively high weight to produce a clear effect on motion continuity / slow-motion tendency / missing transitions / prompt-following. 单独使用时,需要相对较高的权重,才能比较明显地影响动作连续性 / 慢动作倾向 / 动作断链 / 提示词遵循度。 0.5–0.70.5–0.7 Because Combat already provides strong motion speed and impact, Motion Continuity Repair only needs a medium weight to correct continuity / coordination / action logic without unnecessarily taking over the motion. 因为 Combat 已经提供了较强的动作速度和打击感,此时 Motion Continuity Repair 只需要中等权重来修正动作衔接 / 协调性 / 动作逻辑,避免过度接管整个动态。 More complex sequences can gradually increase the Stage 1 weight, but very high weights may begin to reshape the entire motion logic / affect visual style / or alter audio characteristics. 动作越复杂可以逐渐提高一采权重,但过高时可能开始重构整段动作逻辑 / 改变画面表现 / 影响声音特征。 Stage 2: around 0.2–0.3 is recommended. 二采建议约 0.2–0.3。 Higher Stage 2 weights may over-stabilize the sequence and reduce motion intensity / speed / impact. 二采权重过高可能会把动作收得太稳,从而削弱动作幅度 / 速度感 / 冲击力。 最简单地说就是: 单独使用:≈0.9 / 搭配 Combat:0.5–0.7 / 二采:0.2–0.3 这个才是目前测试出来的正确推荐方式。 Although it is a general-purpose LoRA, it works especially well together with a Combat LoRA. Combat provides speed and impact / Motion Continuity Repair focuses on continuity, coordination, prompt-following and interaction logic. 虽然它是一个通用 LoRA,但目前非常推荐与 Combat LoRA 搭配使用。Combat 负责速度与打击感 / Motion Continuity Repair 负责动作衔接、协调性、提示词遵循与交互逻辑。 Trigger Word / 触发词 Trigger word: bunny_crisp_motion — optional. It mainly serves as an explicit identifier for the LoRA and is not required for normal use. 触发词:bunny_crisp_motion —— 可选。它主要用于明确标识这个 LoRA,正常使用时并不是必须填写。 You can test with or without the trigger word depending on your workflow and prompt. 你可以根据自己的工作流和 Prompt,自行测试是否加入触发词。 What does it do? / 它主要做什么? Ever had a great clip ruined by less than one second of strange motion? A sudden slowdown / a missing action / an awkward limb trajectory / a broken interaction can easily make an otherwise usable clip unusable. 你是否遇到过这种情况:一段视频明明整体很好,却偏偏有不到 1 秒突然慢下来 / 漏掉一个动作 / 肢体轨迹变得诡异 / 交互突然断掉,最后整段素材因此报废? This LoRA is designed around repairing those moments rather than simply making every motion stronger. 这个 LoRA 的重点就是修复这些“突然掉链子”的瞬间,而不是粗暴地把所有动作统一增强。 Current A/B testing shows improvements in unwanted slow-motion behavior / missing motion transitions / abnormal limb trajectories / action coordination / prompt-following / interaction continuity / and some mismatched motion-feedback events. 目前 A/B 测试中已经观察到对 异常慢动作 / 动作断链 / 诡异肢体轨迹 / 动作协调性 / 提示词遵循度 / 交互连续性 / 部分错误动作反馈 的改善。 When H3 already handles a scene well, the difference may be very small / when the base model begins to struggle, the difference can become much more obvious. 当 H3 本来就很擅长某段动态时,开启前后的区别可能非常小 / 当底模开始掉速、断动作或失去逻辑时,差异反而会明显很多。 It doesn't simply make H3's strengths stronger — it tries to keep its weaker motion states from falling apart. 它不是单纯让 H3 的强项更强,而是尽量不让 H3 在弱项动态里突然掉下去。 Current A/B Results / 当前对比表现 Different weights can progressively change motion intensity and eventually begin reshaping the overall motion rhythm. 随着一采权重提高,可以逐步改变原本偏弱的动态表现,高权重下甚至会开始重构整体动作节奏。 In complex action sequences, the LoRA can reconnect missing actions and reduce strange limb paths, making the motion chain more complete. 在复杂连续动作中,它可以补回部分没有连接上的动作,并减少诡异肢体轨迹,让动作链更加完整。 In scenes where even a Combat LoRA cannot fully suppress H3's slow-motion tendency, Motion Continuity Repair can further pull the sequence back toward a faster real-time motion state. 在一些连 Combat LoRA 都无法完全压住慢动作倾向的场景中,Motion Continuity Repair 还能进一步把动作往正常高速节奏拉回。 It can also improve action-prompt adherence, turning sequences that originally move randomly into results that follow the intended action order more closely. 它也可以提升动作提示词遵循度,让原本只是胡乱运动的结果,更容易按照 Prompt 中指定的动作顺序执行。 Weapon Interaction / 武器交互 A portion of weapon finisher / execution animation data was included in the dataset. 训练数据中还加入了一部分武器处决 / 终结动画。 Current testing shows some improvement in weapon disappearance / clipping / unstable fast swings / incomplete hit reactions / and certain
3,160 downloads · 229 likes on Civitai
SHA256
EDFA0E2858CAEBA7BE9108883F61DA8525D3C58AF998C94AC7A449D9603D7574
Ranking badge
Updated Oct 22, 2026
BUNNY General Motion Continuity Repair on ModelTrends.ai: #121 Hottest AI model
Made this model? Show this badge on your page.
Paste it into your Hugging Face card or Civitai description.
Also ranked: #3 Hottest MiniMax H3 model · #3 Hottest MiniMax H3 LoRA
Not this creator?
BUNNY General Motion Continuity Repair Popularity Trend
Heat score since 28 Sep 2026
Heat score
8,663°
This week
▲
19.3%
8,628° then, 8,663° now
This month
n/a
Wasn't released/tracked
Past year
n/a
Wasn't released/tracked
BUNNY General Motion Continuity Repair: Heat score since 28 Sep 20267,000°7,466°7,931°8,397°8,863°29 Sep3 Oct8 Oct12 Oct17 Oct22 Oct8,663°ModelTrends.ai
7,000°
Floor
The scale runs from 0° to 9,999°. Everything below 7,000° is kept for models no source reports on yet, which carry a stand-in number there so lists stay complete. 7,000° is where measured data begins: a model with real data but almost no downloads sits at the floor, and the measurements fill the degrees above it. The floor says nothing about the model; only the degrees above it do, and they are what separates one model from another.
+1,070°
Civitai downloads
3,160 downloads of this exact version on Civitai. Downloads are the heaviest ingredient and are counted on a log scale: ten times the downloads adds the same 300°, so a model with a million downloads is not a thousand times hotter than one with a thousand. The six download sources share that one term, split between them in proportion to their counts.
+354°
Likes
229 likes and thumbs up across the sources. A like weighs half a download, on the same log scale.
+139°
Adoption rate
About 70 downloads a day, measured between two readings of the sources, never guessed from a lifetime average. Speed weighs a quarter of the download term, so a fresh surge shows without letting a one-day spike outrank years of use.
+100°
Family boost
+100° because this is a LoRA of a family this site is nudging forward right now. It is a bounded nudge inside the score and is listed here openly; it can never hide a model or outweigh a thousandfold difference in downloads.
Show detailed breakdown
Every number on this site can be checked. This model's heat score is 8,663°, and here it is built one ingredient at a time, with the running total after each: the floor, the degrees each source adds, the bonus for the kind of model and what the age of its weights takes away. The colors are the bands of the chart above. The glossary at the end defines every term.
  1. 1
    Start at the floor.
    Why 7,000° and not zero: the scale runs from 0° to 9,999°, and the part below 7,000° is reserved for models no source reports on yet, which carry a stand-in number there so the lists stay complete and never outrank a model with data. 7,000° is where measured data begins. A model with real data but almost no downloads sits here; everything that follows is measurement, and the room above the floor holds ten tenfold steps in downloads at 300° each. The floor is the same for every model, so it decides nothing; the degrees above it do.
    7,000°
    running total
  2. 2
    Add the downloads.
    3,160 download-equivalents in all, from Civitai downloads. Ten times the downloads adds the same 300°, so size matters but never runs away with the score.
    +1,070°
    8,070°
    running total
  3. 3
    Add the likes.
    229 likes and thumbs up, at half the weight of a download.
    +354°
    8,424°
    running total
  4. 4
    Add the adoption rate.
    About 70 downloads a day right now, measured between two readings, at a quarter of the weight.
    +139°
    8,563°
    running total
  5. 5
    Family boost.
    An open, bounded nudge for LoRAs of a family this site is pushing forward right now.
    +100°
    8,663°
    running total
  6. =
    The heat score.
    Where this model stands on every list of this site. Nothing else adjusts it.
    8,663°
Glossary: the terms used here
Heat score
One number for how much a model is used and wanted right now, from 0° to 9,999°. Every list on this site that says hottest is sorted by it and nothing else.
Floor
The 7,000° every model with real data starts from. The scale runs from 0° to 9,999°; the part below the floor is reserved for stand-ins (models without data), so that a model with data always ranks above one without. The floor is the same for everyone and decides nothing; the degrees above it are the measurement.
Download-equivalents
Everything the score treats as a download, added up: downloads themselves, each torrent seeder as 1,000, each AI Horde image and each run on a generation site as one.
Log scale
Counting by tens, not by ones: going from 1,000 to 10,000 downloads adds the same degrees as going from 10,000 to 100,000. It keeps giants from crowding out everything else.
Likes
Thumbs up and likes from every source, worth half a download each on the log scale.
Adoption rate
How many downloads a day a model gets right now, from the difference between two readings of a source. A quarter of the weight of the download term.
Measured and estimated
Measured means two readings of a source exist and the difference is real. Estimated means only today's total is known and the past is spread over the model's life; the site says which one it is showing.
Seeder
Someone who keeps a copy of the weights online for others to download over a torrent, or a hosted mirror doing the same. Each one counts as 1,000 downloads.
AI Horde
A volunteer network that generates images for free on donated computers. The number of images it has made with a model is public, and each counts as one download.
Generation sites
Sites that host their own copies of a model and let people generate with it, such as PixAI, SeaArt and Yodayo. Their downloads count like Civitai's and each generation run as one download.
OpenRouter tokens
A text model served through an API, such as Claude or GPT, has no weights to download, so its usage is measured in tokens: the pieces of text it read and wrote through OpenRouter, a gateway many apps use, over the last 30 days. Every 10,000,000 tokens count as one download. Source: OpenRouter (openrouter.ai/rankings), CC BY 4.0.
Full model (checkpoint)
The weights themselves, the file that generates on its own. It gets a bonus because add-ons depend on it.
LoRA and embedding
Small add-on files that steer someone else's weights toward a style, a character or a concept. They cannot run alone.
Family and weights
The base a model is built on, such as SD 1.5, SDXL, Flux or Wan. Every model inherits the release date of its family's weights.
Age
Degrees taken off for how long ago the family's weights came out: 150° a year, up to 750°. It keeps old downloads that had years to pile up from freezing the top of the list.
Family boost
A small, bounded nudge for LoRAs of a family this site is pushing forward right now, always listed in the breakdown.
Stand-in score
A stable number below the floor given to a model no source reports on yet, so lists stay complete. It never ranks above a model with data.
New! (grace period)
A model less than 7 days old shows New! instead of a score and holds no place in the rankings. Its score is computed from day one, but a model one day old grows faster than anything, so without the wait the week's #1 would always be whatever came out yesterday rather than what held up for a week.
Trend
How a model's growth over a period compares with the typical model's growth over the same period. Faster than typical reads as a rise, slower as a fall.
Typical model
The middle of the catalog: half the models with a figure grew faster over the period, half slower.
How this model's popularity has moved against the rest of the catalog: its growth in downloads, likes and usage over each period, measured against the typical model's growth over the same period, with its heat score at the time. Above the typical growth reads as a rise, below it as a fall. The past is reconstructed from today's totals, the model's first day and its measured download speed where two readings exist; a model that did not exist at the start of a period shows n/a.
Example renders
Prompts and settings shared by the people who made these renders on Civitai. Pick a render to see how it was made.
No prompt was shared with this render.
No prompt was shared with this render.
Render by tsolful on
Civitai
·
Prompt on BunchaPrompts
Flag
Settings
Steps
8
Guidance
1
Size
1584x1200
Prompt
integrated_multimodal_description: [Shot 1] 00:00.000–00:03.000. Live-action, gritty dark fantasy, overcast grey daylight in a foggy cypress swamp draped with Spanish moss. Starting from the opening frame: the bearded knight in dented, rust-streaked plate armor, an open-faced steel helmet, and a torn green tabard stands knee-deep in murky water, swinging his longsword with a round shield on his arm, as a three-headed hydra with grey-green scales and spiked crests rears up out of the water in front of him, all three jaws open and dripping. A rickety wooden dock with a lit hanging lantern stands to the right. The camera tracks in fast at water level. At 00:00.600 his blade slams into the left head's neck, spraying water and the head recoils with a shriek. At 00:01.800 the center head strikes straight down at him; he throws up his shield and the jaws clamp onto its rim, wrenching him sideways in a burst of spray. [Shot 2] 00:03.000–00:05.500. Hard cut to a low shot skimming the water surface. The right head snakes low and fast across the swamp, jaws snapping at the small green frog; at 00:03.800 the frog leaps onto the rusted helmet half-sunk in the water, and at 00:04.600 springs again onto the dock planks just as the jaws crunch shut on empty air, sending the half-sunk helmet spinning away. [Shot 3] 00:05.500–00:08.000. Hard cut to a medium shot on the knight. He rips his shield free of the center head's jaws, spins, and at 00:06.200 smashes the shield edge into its snout, knocking the head aside. The left head lunges back in; he ducks under it, water churning around his armored legs. The knight with a rough, breathless voice (S1) snarls: <d>[English] Come on, then, you ugly beast!</d> [Shot 4] 00:08.000–00:10.000. Hard cut to a wide shot of the swamp. All three heads rear up together, roaring. The knight charges forward through the water, sword raised high, kicking up spray with every stride. At 00:09.200 the right head whips sideways and smashes through a dock post, the lantern swinging wildly and wooden splinters flying, as the knight closes in on the hydra's body. overall_soundscape: Sloshing water, dripping moss, and croaking swamp insects throughout. A heavy steel chop and a piercing reptilian shriek at 00:00.600, a crunching clang of teeth on the shield at 00:01.800, the frog's croak and small splashes, a hollow metallic ping as the helmet is knocked, a wet snap of empty jaws, a dull metallic smash of shield on snout at 00:06.200, the knight's heavy breathing, triple overlapping roars, splashing strides, and a splintering crack of timber with the lantern's chain rattling at 00:09.200. non_diegetic_music: A gritty, fast-tempo orchestral score with low pounding drums, harsh brass, and tense strings, hitting accents on each strike and building to the final charge.
Render by DivideBiZero on
Civitai
·
Prompt on BunchaPrompts
Flag
Settings
Sampler
Euler
Steps
8
Size
1536x2176
Prompt
or the target video, at 0.00 seconds into the target video, <Picture 1> (from [Shot 1]) is fully referenced. integrated_multimodal_description: [Shot 1] BUNNY, prfight2, prfin1 Static locked-off camera, no zoom, no pan, no cuts. The Headless Horseman, a headless rider in a tattered dark cloak, leather jerkin, and riding boots, sits astride a massive dark horse with glowing orange eyes, a pale flowing mane and tail, and cracks of molten fire glowing through its hide, in a dark misty forest beneath a full moon, holding a flaming jack-o'-lantern aloft in one gloved hand. Throughout, flames lick and roll along the horse's legs and hooves, embers and sparks drift upward, the rider's tattered cloak billows, fallen leaves swirl along the ground, and mist drifts between the bare trees. Early in the scene, the horse rears up high on its hind legs, front hooves kicking through the air trailing fire, and lets out a loud, terrifying whinny. It then stretches its neck forward and breathes a roaring stream of fire from its mouth that lights up the forest. As the horse drops back onto all four hooves, the Horseman draws back his arm and hurls the flaming jack-o'-lantern straight toward the camera; it tumbles forward, grinning and trailing fire and sparks, growing larger until it slams into the lens and bursts into an explosion of flame and embers that fills the frame. Audio: a fearsome horse whinny, pounding hooves, a roaring whoosh of fire, crackling flames, and a fiery explosion. Powerful, dramatic motion, consistent moody lighting, cinematic dark Halloween fantasy, high detail, 15 seconds.
No prompt was shared with this render.
No prompt was shared with this render.
Render by jeh88615 on
Civitai
·
Prompt on BunchaPrompts
Flag
Settings
Sampler
Euler a
Steps
20
Guidance
1
Seed
2184774017
Size
1104x1968
Prompt
integrated_multimodal_description: <Picture 1>[Shot 1] Cinematic, A sumiink painting style animation,Japanese woman,Street dance,slim waist,<Picture 1> A single girl stands in a neon-lit city at night, her electric hair glowing with energy against a graffiti-covered background alive with light. Her sharp, confident eyes lock forward as she strikes a high-energy pose, the holographic katana in her hand shimmering with shifting vaporwave colors. She wears street fashion with sleek techwear elements — layered fabrics, futuristic sneakers, and reflective details that catch the city’s neon glow. Sparks fly around her as if the air itself hums with static charge. The dynamic camera angle enhances the sense of motion and power, while the lighting is dramatic and cinematic, flooding the scene with blues, pinks, and purples. The atmosphere feels alive and electrifying, capturing the perfect fusion of cyberpunk energy, vaporwave style, and modern anime intensity. Finally, a downward slash; the blade strikes the ground, unleashing a burst of flames. overall_soundscape:Street dance music, the sound of knives being swung, the sound of knives scraping against the ground.
No prompt was shared with this render.
Run Locally
Motion_Repair_V2.safetensors
148 MB · SafeTensor · bf16
Downloads are served by Civitai. Some creators ask for a Civitai login first.