大家好！我是模型创作者poakl。

这个Lora是基于Hunyuan1.5底模训练的文生视频Lora。作用是：增强生成视频的真实电影感，提升画面质感、光影层次与胶片氛围。

可以根据需要加入一些镜头描述（如35mm变形镜头、浅景深、手持晃动）、光影描述（如黄金时段侧光、体积光、霓虹反射）、胶片风格描述（如Kodak Vision3颗粒、Fuji Eterna色调、高ISO噪点），详细参考封面实例视频的提示词。

触发词：真实感提升。

建议权重：0.6。

⚠️ 小贴士：

• 权重0.4-0.5适合轻微提升质感，保留原有风格；

• 权重0.6为推荐值，电影感与画面平衡最佳；

• 权重0.7-0.8适合追求强烈胶片电影风格，但可能影响原有色彩。

希望这个Lora能给您的使用带来优化和帮助！欢迎多多使用！多多返图！

---------------------------------------------------------------------------------------------------

Hello everyone! I'm the model creator poakl.

This LoRA is trained on the Hunyuan 1.5 base model for text-to-video generation. Its purpose is to enhance the realistic cinematic quality of generated videos, improving visual texture, lighting depth, and film atmosphere.

You can add lens descriptions (e.g. 35mm anamorphic lens, shallow depth of field, handheld shake), lighting descriptions (e.g. golden hour side light, volumetric rays, neon reflections), and film style descriptions (e.g. Kodak Vision3 grain, Fuji Eterna tones, high ISO noise) as needed. Please refer to the cover example video prompts for detailed guidance.

Trigger word: Realism 真实感提升.

Recommended weight: 0.6.

⚠️ Tips:

• Weight 0.4-0.5 for subtle texture enhancement while preserving the original style;

• Weight 0.6 is the recommended value for the best balance between cinematic feel and image quality;

• Weight 0.7-0.8 for a stronger film grain cinematic look, though it may affect the original color palette.

I hope this LoRA enhances your creative workflow! Feel free to use it and share your results!