Pro52 / Qwen3.5 Edition Anima visual style powered by Qwen3.5 4B's language understanding. 52-block DiT · Qwen3.5 4B Nodes Load Qwen3.5 4B (Anima) select qwen35_4b.safetensors Qwen3.5 Unified Prompt (Anima) plug model, native clip (0.6 - original anima clip), and qwen_35 clip into the node's INPUT provide the prompt* ** note on prompt: This was trained with the Description separator for NL first write your normal meta/artist/quality tags then Description: e.g: masterpiece, best quality, high quality, newest, year 2025, year 2024, Description: A fox-girl jumps over a high fence. Select the anima-3.8b adapter, you can leave strength at 1.0. Use the expanded OUTPUT for positive conditioning. Likewise it is recommended to do the same with negative prompt/conditioning. Note on the native OUTPUT: This is the original qwen 0.6b conditioning Since Anima 3.8B was trained at low res, it is recommended to plug native into the second ksampler/highres pass to avoid artifacts Recommended starting point Resolution: 832x1216 px (or any other 1MP res) Adapter strength: 1.0 CFG: 7–8 Steps: 28–50 Sampler/scheduler: res_multistep + Beta (generally anything with Beta) At adapter_strength: 0, conditioning is exactly native. Increase it while holding the seed fixed. High CFG often makes the learned semantic branch more visible; if the output becomes harsh or characters homogenize, reduce CFG before reducing strength. Prompting Natural language, tags, and hybrids all work. For complex scenes, assign each subject its own sentence: Miku on top right. Teto on lower left. Apple on top left. Nothing on lower right. etc. Try to avoid pronouns when working with multiple characters, e.g: she/he, use explicit names instead; e.g: Instead of 2girls, Miku is sitting near Teto, she is eating an ice-cream. Try: There are two girls in this illustration. Miku from Vocaloid and Teto from Vocaloid. Miku is sitting near Teto, Miku is eating an ice-cream. Made by - LyloGummy https://civitai.com/models/2880480/anima-38b