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.
Prompt
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).
Recommended
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).
Anima 52 LoRA Patch only for v1
https://github.com/bestluner-create/ComfyUI-Anima52-LoRAPatch
Anima 52 LoRA Patch for v2+
https://github.com/bestluner-create/ComfyUI-Anima-Delta-Q35-LoRA.git
Start around 0.65. Do not stack stock Load LoRA on the same old LoRA.
v2
Current version on this page.
This mix uses: Anima-3.8B Anima Aesthetic 2B Aura Rising aniMurr v02
Same text encoder and VAE as v1.0. File: animaDeltaMix38B_v20.safetensors
v3
New bake on the same 52-block 0.6B chassis
Both Gen-1 and Gen-2 use copy-body (adaLN stays on the chassis)
Denser materials and small surface detail vs v2
Same 2B donors — last styled bake on this pair
V4
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.
V5/V6
Used custom donors / Dora
V7
A clean anime style with sharp lines and bright colors.
V8
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.
Credits
Anima Base / Anima-3.8B — original authors
Method: Anima Delta Mix — 2B into 2.9B / 3.8B
License follows Anima (non-commercial). Do not sell the weights.
Description
FAQ
Comments (5)
does this model use the base vanilla diffusion model loader?
Yes it uses the standard diffusion model loading path. The text encoder is native qwen_3_06b_base; the 3.8B model is the 52-block DiT itself.
@bestluner180 Oh cool, the outputs look really cool and i wanted to try the model out.
キャラクターLoraを使うと絵柄がそのキャラクターに寄せられてしまうので
このモデルの絵柄を固定できるLoraが欲しいです
キャラLoRAで絵柄が寄る件、了解です。
このモデル用の「絵柄固定LoRA」は作らない方がいいです。すでにスタイルが入ったチェックポイントに、同じ絵柄のLoRAを重ねると固定ではなくノイズが増えて崩れます。あのやり方は Raw / Base 向けです。
実用的な対処は、モデル側ではなくキャラLoRAから余計な層を切ることです。SDXLの LoRA Block Weight と同じ系統が Anima にもあります。
https://github.com/DanrisiUA/ComfyUI-LoRA-Block-Filter
まずプリセット no_style_leak（style寄りの cross-attn を落とす）
https://github.com/rom0718/Comfyui-Anima-lora-block-weight
https://github.com/Sen-sou/Bobs-Lora-Loader-Anima
キャラLoRAをこのフィルタ経由で、強さは少し下げて試してください。