━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

🔥 RunningHub / RH

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

🎁 New users on RunningHub International can get 1000 RH Coins after registration.

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

Tested hardware: RTX 4070 Ti 12GB VRAM.
实测硬件：RTX 4070 Ti 12GB 显存。

A 10-second H3 video at around 1080P / 2.0MP takes approximately 15 minutes locally on 12GB VRAM.
10秒 H3 视频在约1080P / 2.0MP下，本地12GB显存实测生成时间约 15分钟。

The same 10-second test takes approximately 7 minutes on RunningHub.
同样的10秒测试，在 RunningHub 上大约 7分钟完成

国内用户可以使用：https://www.runninghub.cn/post/2102283144824184833/?inviteCode=cn-v1093

🎬 BUNNY H3 High-Dynamic Two-Pass Latent Upscale Workflow

▶ Start / End Frame One-Click Workflow

▶ Multi-Reference One-Click Workflow

📦Bunny Official RunningHub Model Page

Future FourBunny LoRAs will also be released and updated on RunningHub.

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Highly recommended for complex action: pair this LoRA with BUNNY H3 Conditioning Bridge for better multi-character logic, action ownership, and scene stability.
复杂动作场景强烈推荐搭配 BUNNY H3 Conditioning Bridge，可进一步改善多人互动逻辑、动作归属和场景稳定性。

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

⚠️ IMPORTANT — WORKFLOW REQUIREMENT

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Please use my workflow when testing this LoRA.

This release has been tested and tuned specifically with my workflow. Some third-party workflows or plugin-based workflows may introduce additional blur, motion degradation, sampling differences, or other issues unrelated to the LoRA itself.

I will no longer provide troubleshooting or support for image blur / quality degradation caused by third-party plugin workflows.

请务必使用我提供的工作流测试本 LoRA。

本版本的测试与参数均基于我的工作流完成。部分第三方工作流或插件工作流可能会额外导致画面模糊、动态退化、采样差异等问题，这些问题并非 LoRA 本身导致。

今后对于使用第三方插件工作流所产生的画面模糊、质量下降等问题，我将不再进行排查和回复。

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

⚔️ Weapon Combat V1

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

A MiniMax H3 LoRA focused on high-speed weapon combat, cinematic choreography, weapon interaction, large body movement, weapon flourishes, heavy impacts and execution sequences.

一个专注于 MiniMax H3 高速武器战斗的 LoRA，主要强化影视化武指、武器交互、大幅度身体位移、武器花式、重击反馈与处决动作。

Designed for swords, sabers, spears, polearms, axes, hammers, shields, fantasy weapons and energy blades.

适用于剑、刀、长枪、长柄武器、战斧、战锤、盾牌、奇幻武器以及能量武器等。

Compared with the base model, it improves:

• Faster and more continuous weapon combat

• Better attack → contact → deflection → recovery logic

• Weapon-to-weapon interaction

• Large swings, rotations and spatial repositioning

• Weapon flourishes connected directly into attacks

• Heavy impacts and execution motion

• More cinematic weapon choreography

相比底模，主要强化：

• 更快、更连续的武器战斗

• 更清晰的 攻击 → 接触 → 偏转 → 回收 → 下一击 逻辑

• 更好的武器对武器交互

• 大范围挥击、旋转与空间换位

• 武器花式直接衔接真实攻击

• 更强的重击与处决动作

• 更具影视感的武器武指表现

Additional Note:

This LoRA includes training related to execution-style finishing motion and dismemberment-adjacent action concepts.

those examples are not included in the public preview.

They have been tested privately and are part of the LoRA’s behavior range.

补充说明：

这个 LoRA 包含了处决式终结动作，以及与断肢相邻的动作概念训练。

这部分内容不会出现在公开预览中。

但我已经私下完成测试，它们属于本 LoRA 的能力范围之一。

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

🐇 TRIGGER WORD

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Weapon Combat LoRA Trigger Word:

BUNNY

No typo here — the trigger word really is BUNNY 🐇

武器战斗 LoRA 触发词：

BUNNY

你没看错，真的就是 BUNNY 🐇

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

⚙️ RECOMMENDED STRENGTH

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Recommended Strength:

0.7–0.9 to start.

You can try 1.0 for stronger motion, but higher strength may occasionally increase excessive full-body rotation.

推荐权重：

建议从 0.7–0.9 开始。

需要更强动作表现时可以尝试 1.0，但较高权重偶尔会增加人物整体旋转倾向。

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

⚠️ V1 NOTES

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Weapon stability and combat logic are significantly improved, but weapon deformation, disappearance and contact errors can still occur in extremely complex scenes or crowded fights.

武器稳定性和战斗逻辑已经明显改善，但在极复杂场景、多人混战或高密度动作下，仍可能出现武器变形、消失或接触错误。

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

🎁 BONUS

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Included:

“H3 Weapon Combat Choreography Director”

Give the TXT directly to ChatGPT / Claude / Gemini / Grok, then tell it your characters, weapons, scene, duration and desired action style.

The template includes choreography design, camera movement, spatial movement, weapon techniques, execution design, transitions, sound design, and good/bad prompt examples.

附件包含：

《H3 Weapon Combat Choreography Director 》

可以直接把 TXT 文件交给 ChatGPT / Claude / Gemini / Grok，再告诉 AI 你的人物、武器、场景、时长以及想要的动作风格。

模板已经包含武器武指、运镜、空间移动、武器技法、处决、转场、音效设计，以及“好看 / 难看”提示词对照示例。

即使 AI 完全没有之前的聊天记忆，也可以直接使用。