V1.1

Minor bug fixed in the lora attach area. Add strength option for every lora.

更正lora 區的小錯誤. 為每一個Lora補回權重設定欄.

The purpose of this LoRA setting is that when generating a QuadView, if the original image was created using a character/outfit/style LoRA, we also need to add the LoRA here to make up for the base model's limitations. I sincerely apologize for forgetting to include the weight field.🙏

這個Lora設定目的, 在生成QuadView時, 假如原圖是用服裝/角色/style Lora生成, 這兒亦需加載Lora來彌補原model的不足, 忘記了加權重欄, 萬分抱歉.🙏

(有中文)Why Does AI Always Give You a Close-Up? How to Control Framing with a Four-Stage Ladder 📷✨

https://civarchive.com/articles/34198/why-does-ai-always-give-you-a-close-up-how-to-control-framing-with-a-four-stage-ladder

Minimax H3 simple workflow for low VRAM:

https://civarchive.com/models/2862009/minimax-h3-multi-ref-workflow-for-low-vram

VoxCPM simple workflow for low VRAM (Audio dialogue generation model)

https://civarchive.com/models/2865730/voxcpm2-simple-workflow-for-low-v-ram

VAE and Text Encoder on the right side is not require, you might use the current use model instead.

only the Lora and node of edit are required.

右方推薦的VAE和Text Encoder不是必須的, 可以用目前使用的代替.

This is a simple workflow for creating QuadViews. Detailed instructions are provided in the article below, so I won't repeat them here.

簡單的workflow 用來製作QuadView, 詳細會在下方文章說明, 這兒就不重覆了

https://civarchive.com/articles/34064/how-to-keep-character-appearance-consistent-in-every-shot-minimax-h3

All credit belongs to lora creator and node creator:👏