<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/32753658-669e-4a96-a59d-b880604ac2f9/original=true/32753658-669e-4a96-a59d-b880604ac2f9.jpeg" />IMPORTANT. Read this first. This checkpoint is not built for Qwen3.5-4B and not for the expanded adapter.52-block DiT, but text conditioning is native qwen_3_06b_base only. Insert blocks after the frozen 28-stem were prepared for that CLIP. Qwen3.5 will look like stock Pro52, not this mix.Do not use qwen35_4b or Anima-3.8B-expanded_adapter. Delta Mix Checkpoint merge on Anima-3.8B (52-block). Anime and illustration first, with extra room for kemonomimi / light furry. Use it like a normal Anima model.Text encoder: qwen_3_06b (native Anima CLIP). Do not swap in Qwen3.5-4B or an expanded adapter.VAE: official Anima / Qwen-Image VAE.<h3 id="prompt">Prompt</h3>Positive starter: anime, masterpiece, best quality, ultra detailed, absurdres, sharp focus, intricate details, cinematic dramatic lighting, strong chiaroscuro, Then the scene in tags or natural language, Anima-style. Rating tag if you use one (safe / sensitive / explicit).<h3 id="recommended">Recommended</h3>Steps: 30-40 CFG: 4.0–5.5 Sampler: Euler ancestral/dpmpp_2m Scheduler: normal/sgm_uniform Resolution: same as Anima-3.8B (1024-class and up, as VRAM allows).<h3 id="anima-52-lora-patch-only-for-v1https:github.combestluner-createcomfyui-anima52-lorapatchanima-52-lora-patch-for-v2+https:github.combestluner-createcomfyui-anima-delta-q35-lora.git">Anima 52 LoRA Patch only for v1 <a target="_blank" rel="ugc" href="https://github.com/bestluner-create/ComfyUI-Anima52-LoRAPatch">https://github.com/bestluner-create/ComfyUI-Anima52-LoRAPatch</a> Anima 52 LoRA Patch for v2+ <a target="_blank" rel="ugc" href="https://github.com/bestluner-create/ComfyUI-Anima-Delta-Q35-LoRA.git">https://github.com/bestluner-create/ComfyUI-Anima-Delta-Q35-LoRA.git</a></h3><h3 id="start-around-0.65.-do-not-stack-stock-load-lora-on-the-same-old-lora.v2">Start around 0.65. Do not stack stock Load LoRA on the same old LoRA. v2</h3>Current version on this page.This mix uses: Anima-3.8B Anima Aesthetic 2B Aura Rising aniMurr v02Same text encoder and VAE as v1.0. File: animaDeltaMix38B_v20.safetensors<h3 id="v3">v3</h3><h3 id="new-bake-on-the-same-52-block-0.6b-chassis">New bake on the same 52-block 0.6B chassis</h3>Both Gen-1 and Gen-2 use copy-body (adaLN stays on the chassis)Denser materials and small surface detail vs v2Same 2B donors — last styled bake on this pair<h3 id="v4">V4</h3>Custom models are used for both generations, including the stem. This specific model focuses on delivering a high-quality output with maximum detail and fidelity. I hope you enjoy it.<h3 id="v5v6">V5/V6</h3>Used custom donors / Dora<h3 id="v7">V7 </h3>A clean anime style with sharp lines and bright colors. <h3 id="v8">V8</h3>A new 3.8 assembly with different donors. The main focus is on background filling and overall frame detail.It’s still an early version: it breaks hands and fingers more often, anatomy can suffer, and background artifacts can appear.<h3 id="credits">Credits</h3><ul><li>Anima Base / Anima-3.8B — original authors</li><li>Method: Anima <a target="_blank" rel="ugc" href="https://civarchive.com/models/2908374/anima-delta-mix-2b-into-29b-38b">Delta Mix — 2B into 2.9B / 3.8B</a></li></ul>License follows Anima (non-commercial). Do not sell the weights.

Anima Delta Mix 3.8B Qwen 0.6B

v3.0<ul><li>New bake on the same 52-block 0.6B chassis</li><li>Both Gen-1 and Gen-2 use copy-body (adaLN stays on the chassis)</li><li>Denser materials and small surface detail vs v2</li><li>Same 2B donors — last styled bake on this pair</li></ul>Sampler unchanged: 40 steps, CFG 6, scheduler normal. Same-model 2nd pass at 1.25 if you want a tighter frame.

Files with SHA256: 9ad9a20565e3d0ca466b50b59cf06d52753f89c08790229af1c14264cebfa071