Anime Styler (Advanced)

The same 49 anime-genre presets as the plain Anime Styler - Chibi Style, Mecha Anime, Magical Girl Transformation, Shonen Battle - but with the two-box, six-output treatment. AnimeStylerAdvanced from aegis72/comfyui-styles-all is aimed at SDXL setups that want a separate global prompt for the OpenCLIP-G encoder and a local prompt for CLIP-L, plus fine-grained control over where the negative lands.

The inputs

• text_positive_g / text_positive_l - your two positive boxes.
• text_negative - your negative; the style's negative gets appended.
• Anime - the 49-option genre dropdown.
• negative_prompt_to - Both (default), G only, or L only.
• log_prompt - on by default; prints the before/after to the console.

Outputs: text_positive_g, text_positive_l, text_positive (combined), text_negative_g, text_negative_l, text_negative.

The catch with these templates

Two things to know before you build around this node. First, the anime templates are single-part wrappers ((Chibi anime style) {prompt}), so there's no . separator for the G/L machinery to actually split. Second, the port swaps the two boxes: your text_positive_l text is what gets styled and comes out the text_positive_g output, while text_positive_g passes through unstyled to text_positive_l.

The path of least resistance:

• Put your prompt in text_positive_l and read the combined text_positive output into a CLIPTextEncode (or CLIPTextEncodeSDXL on SDXL).
• Or skip the confusion entirely and use the plain Anime Styler - one box, two outputs, same 49 styles.

Installing it

ComfyUI Manager (search ComfyUI-styles-all), or:

cd ComfyUI/custom_nodes
git clone https://github.com/aegis72/comfyui-styles-all

Then restart ComfyUI. No requirements.txt, no models.

Where people get burned

• The G/L swap is the one that bites - if your styled output is missing the subject, that's the cause.
• Restart after editing data/Anime/Anime.json for new options.
• Anime checkpoints love it, Flux doesn't. These genre keywords were built for SDXL's CLIP encoders; on anime SDXL-lineage models (Illustrious, NoobAI, Pony) the negatives actually bite at CFG 4–9, but LLM-encoded models mostly ignore them.