💡 Support & Official Links Note: If you enjoy my model content, please support me with a donation on my Ko‑fi page. Your support helps me continue developing tools and models for everyone. Download links for the models will be available on my Ko‑fi page for users who support the development of my models, allowing them to be used locally. Thank you for your attention. 👉 https://ko-fi.com/jaceraimi This model is officially presented by Jace Raimi on the following pages: https://tensorhub.art/models/992095100488169536 https://tensor.art/models/992095100488169536 https://www.seaart.ai/es/models/detail/d78ivide878c739gkp8g https://tungsten.run/model/7nwHdDucqa?model_version=Daoixf7xcY https://yodayo.com/models/06aa64d1-5e56-49cb-b0b3-096b21bf2d07?index=1&modelversion=fb37b6a9-c107-412e-96a4-c1bf878d965c https://moescape.ai/models/06aa64d1-5e56-49cb-b0b3-096b21bf2d07?index=1&modelversion=fb37b6a9-c107-412e-96a4-c1bf878d965c 📜 Changelog — FWAI_Anima_Magic_v2.2.6_(R5) 🎨 Style and Tags Enhanced anime details: refined facial features, improved anatomy, and coherent poses. Smooth effect support: more fluid and natural rendering. Combat dataset improvements: optimized for dynamic action scenes and positions. Popular anime style: aesthetics inspired by mainstream anime trends. Better integration of Danbooru‑style tags, enabling richer and more precise prompt control. Illustrious‑inspired soft style: compatible with existing prompts, offering smoother anime aesthetics. ✨ Quality Functions & Style Triggers Activation triggers: masterpiece, best_quality, absurdres, anime_coloring, 2D, anime_screencap, official_artstyle, traditional_media, cinematic_lighting. Speech bubbles and signs supported in prompts. Extended resolution support up to 2000 × 2000, enabling 1080p outputs (functional only in ComfyUI). Expanded Danbooru tag coverage for natural English prompt triggers. 🛠️ Technical Improvements Optimized inference for multi‑tag prompts. Reduced artifacts in complex backgrounds and dynamic lighting scenes. Balanced sharpness and smoothness. Stable for long prompts and advanced tag combinations. ⚙️ Core Architecture Built on Anima: https://huggingface.co/circlestone-labs/Anima Uses anima-base-v1.0.safetensors as foundation, replacing the previous anima-preview3-base.safetensors. Fine‑tuned with anime dataset for improved fidelity. Requires VAE and Text Encoder as separate files: Text Encoder (Qwen3) → https://huggingface.co/circlestone-labs/Anima/tree/main/split_files/text_encoders Qwen‑Image VAE → https://huggingface.co/circlestone-labs/Anima/tree/main/split_files/vae 🚧 Notes Fine‑tuning is still being tested, but this version already produces a smoother, more detailed anime style compared to experimental builds. Designed as a stable base for LoRA training.