凤希AI - 分镜生成器V2 - 进Q群与更多的群友学习：775649071

FxaiStoryBoardV2 ("凤希AI - 分镜生成器V2") is the "V2" of the storyboard generator, and here's the thing nobody tells you: it is the exact same code as V1. In the source, FxaiStoryBoardV2 = FxaiStoryBoard - it's an alias, a rename in the display-name table. Same inputs, same outputs, same behavior, same everything. So this article is mostly about how to recognize that and not waste an afternoon comparing them.

What it does

Everything I said about FxaiStoryBoard applies word for word. You feed it a JSON list of shots in lines_data, and it returns:

• 分镜数据 - LIST of your parsed shots.
• 循环次数 - INT, how many loop iterations it takes to walk the range.
• 开始索引 - INT, where the walk starts (from the 1-based 开始序号).

Optional inputs: 开始序号, 结束序号, and 刷新标记. Same JSON-parses-or-empty-list behavior, same 1-based start index, same clamp-to-minimum-1 loop count.

Which one should you use?

Flip a coin, seriously. Pick V2 if you want your workflow to look current in case the author later changes V2 into something different - a "buy the newer version number" insurance policy. Pick V1 if you don't care about that and want the name that matches the pack's docs. There is no functional difference to discover, and anyone telling you V2 generates better storyboards is selling you something.

This is also a handy lesson in the pack's update habits: several nodes got a V2/V3/V4 that are actually different (the video generators are a good example - see those articles), but this one is a pure relabel. When you see a "V2" in this pack, it's worth 30 seconds to check whether the internals changed. Here, they didn't.

Installing it

Same pack, same one-line install:

cd ComfyUI/custom_nodes
git clone https://github.com/fxai666/fxai-toolkit

Restart ComfyUI (or ComfyUI Manager → search "fxai-toolkit"). When you search the node list you'll see both 分镜生成器 and 分镜生成器V2, and now you know the truth about why they're both there.