# [FWAI Easy Workflow Advanced Anima v2.2] – Anima Workflow

Author: JaceRaimi

## Added

- Support for FWAI ANIMA Magic model with Quality Functions & Style Triggers (`masterpiece`, `best_quality`, `absurdres`, `anime_coloring`, `2D`, `anime_screencap`, `official_artstyle`, `traditional_media`, `cinematic_lighting`).

- Extended resolution up to 2000 × 2000 (1080p outputs), functional in ComfyUI.

- Speech bubble and sign prompts supported in English:

- She has a speech bubble that says "insert text here."

- He's holding a rectangular sign that reads "FWAI ANIMA MAGIC".

- Save Image modes:

- Raw Mode → saves the initial unprocessed output.

- Task Completed with Upscaler → saves the final enhanced image.

- Intermediate stages (Face Detailer, Hand Detailer) for refinement comparison.

## Changed

- Removed RGB Template and Regional Prompt Channels.

- Workflow simplified into a linear pipeline:

- General prompt (Positive/Negative).

- Refiners (Face/Hand Detailer).

- Save Image modes.

- Layout refined into a minimalist, modular design.

- SAM + YOLOv integration maintained for precise face and hand detection.

## How to Use

1. Select the Anima Checkpoint, Text Encoder (`qwen_3_06b_base.safetensors`), and VAE (`qwen_image_vae.safetensors`).

2. Write the full prompt in Positive/Negative fields (global context).

3. Optionally activate up to 5 LoRA modules.

4. Configure sampler, steps, and refiner.

5. Activate YOLOv Face/Hand if desired.

6. Select a custom upscale model.

7. Run the workflow and save images at different stages (Raw, Face Detailer, Hand Detailer, Final Upscaler).

8. Refer to Flow Notes (ES/EN) for contextual guidance.