Experimental Continuous Fight Interaction LoRA 实验性连续战斗交互增强 LoRAMy goal is simple: push H3 from “can fight” to “knows how to fight.” 我的目标很简单：让 H3 从“能打”，变成真正“会打”。🚀 Run It on RunningHub🎁 New users on RunningHub International can get 1000 RH Coins after registration.<a target="_blank" rel="ugc" href="https://www.runninghub.ai/zh-cn/post/2102271264323829761/?inviteCode=rh-v1679">NEW Workflow name: BUNNY_SelfLift_12GB Direct 1080P_High Motion + LoRA Sharp_Hybrid Workflow 工作流名称：BUNNY_SelfLift_12GB直出1080P_高动态+LORA不模糊_混合工作流</a>国内用户可以使用：<a target="_blank" rel="ugc" href="https://www.runninghub.cn/post/2102283144824184833/?inviteCode=cn-v1093">https://www.runninghub.cn/post/2102283144824184833/?inviteCode=cn-v1093</a><a target="_blank" rel="ugc" href="https://www.runninghub.ai/zh-cn/post/2095927034992533505/?inviteCode=rh-v1679">🎬 BUNNY H3 High-Dynamic Two-Pass Latent Upscale Workflow</a><a target="_blank" rel="ugc" href="https://www.runninghub.ai/zh-cn/post/2093321887345819650/?inviteCode=rh-v1679">▶ Start / End Frame One-Click Workflow</a><a target="_blank" rel="ugc" href="https://www.runninghub.ai/zh-cn/post/2093328847935950849/?inviteCode=rh-v1679">▶ Multi-Reference One-Click Workflow</a><a target="_blank" rel="ugc" href="https://www.runninghub.ai/zh-cn/user-center/2092843565587107841/model?inviteCode=rh-v1679">📦Bunny Official RunningHub Model Page</a><hr /><h2 id="what-is-this">What is this?</h2><h2>这是干什么的？</h2>Fight Flow V1 does not mainly teach new moves. It explores whether H3 can learn how two or more characters should actually fight together. Fight Flow V1 主要不是教 H3 更多招式，而是在实验：能不能让两个或多个人真正围绕同一场战斗进行互动。Initiation → response → contact → consequence → initiative shift → next action. 发起 → 回应 → 接触 → 结果 → 主动权转换 → 下一动作继承。The previous action should create the starting state of the next one, instead of constantly resetting back to a neutral pose. 上一招产生的位置、重心、距离和姿态，应该成为下一招的起点，而不是打一轮就重新摆架势。<hr /><h2 id="d-opsd-trained">D-OPSD Trained</h2><h2 id="d-opsd">D-OPSD 训练版本</h2>Unlike regular LoRA training that mainly relies on captions, D-OPSD allows the Teacher to directly reference the target training video. 不同于主要依赖 Caption 的普通 LoRA，D-OPSD 训练时 Teacher 可以直接参考目标训练视频。This helps learn motion relationships that are difficult to fully describe with text: contact, trajectory, forced movement, weapon interaction, initiative changes and state transitions. 这样可以学习很多文字难以完整描述的内容，例如接触、轨迹、被动位移、武器交互、主动权变化和动作状态继承。This V1 was trained entirely on RTX 6000D, with VRAM peaks around 60GB and roughly 2–3× my usual LoRA training time. 这一版全程使用 RTX 6000D 训练，显存峰值约 60GB，整体训练时间大约是我平常 LoRA 的 2～3 倍。<hr /><h2 id="what-does-it-improve">What does it improve?</h2><h2>它主要增强什么？</h2>Better attack-response causality, more natural initiative changes, stronger state continuity, and fewer “attack → reset → attack again” sequences. 攻击与回应的因果更完整，主动权转换更自然，状态继承更明显，也更少出现“打一招 → 重置 → 再打一招”。In multi-person fights, one opponent can leave the exchange while another enters, instead of everyone simply performing independent actions at the same time. 在多人战中，当前敌人退出、主角换位、下一名敌人接入的衔接也会更加自然，而不是所有人同时各打各的。Even with only a short prompt + trigger, the LoRA can already organize fairly complete combat exchanges by itself. 目前测试中，即使只使用一句简单 Prompt + 触发词，也已经能够自行组织出比较完整的连续战斗。<hr /><h2 id="it-is-not-a-speed-lora">It is NOT a speed LoRA</h2><h2 id="&quot;-lora&quot;">它不是“加速 LoRA”</h2>The improvement is not simply “make both characters move faster.” It is mainly about how motion is organized and passed from one action into the next. 它并不是简单把两个人加速，让画面看起来更激烈；真正增强的是动作如何被组织，以及上一动作如何把状态传递给下一动作。<hr /><h2 id="three-interaction-triggers">Three interaction triggers</h2><h2>三大交互触发词</h2><ul><li><code>Unarmed combat interaction.</code> — Use when punches, kicks, grappling, throws or body control dominate the fight. <code>Unarmed combat interaction.</code> —— 拳脚、摔投、身体控制等徒手交互主导时使用。</li><li><code>Melee weapon interaction.</code> — Use when blades, staffs, spears or other melee weapon lines dominate the exchange. <code>Melee weapon interaction.</code> —— 刀、剑、棍、枪等近战武器的攻击线、格挡和兵器接触主导时使用。</li><li><code>Gun-fu interaction.</code> — Use when muzzle control, firearm handling, close-range shooting or gun-hand interaction dominates the fight. <code>Gun-fu interaction.</code> —— 枪口控制、控枪、近距离射击和枪械近战交互主导时使用。</li></ul>Choose the trigger by the dominant interaction logic, not simply by what equipment appears on screen. 判断时看整段战斗由什么交互逻辑主导，而不是单纯看角色手里拿了什么。Body / fists → Unarmed | Melee weapon line → Melee Weapon | Muzzle / firearm control → Gun-fu 身体 / 拳脚主导 → 徒手｜近战武器线主导 → Weapon｜枪口与枪械控制主导 → GunFu<hr /><h2 id="recommended-use">Recommended use</h2><h2>推荐使用方式</h2>Fight Flow works as an interaction layer together with Combat, Weapon and Motion Repair LoRAs. Fight Flow 更适合作为交互层，与 Combat、Weapon、GunFu、Motion Repair 等 LoRA 组合使用。Specialized LoRAs decide what moves the characters know. Fight Flow helps those moves become one continuous fight. 专项 LoRA 决定角色会什么动作，Fight Flow 负责让这些动作真正连接成一场战斗。Current tested weight: 0.65 in the first pass / 0.25 in the second pass. 当前测试权重：一采 0.65 / 二采 0.25。<hr />Before: two characters are both fighting. 以前：两个人都在打。Fight Flow V1: they are fighting the same fight. Fight Flow V1：两个人开始真正打同一场架。

Files with SHA256: 579e8eca67b6d70f4db94b2db67cd938751d576ee3d72aba875dae3d911f7398