Skip to content
Other

Anima

2B text-to-image для аниме и иллюстраций

2 min read

Anima — это text-to-image модель на 2 млрд параметров от CircleStone Labs и Comfy Org. Она заточена под аниме-концепты, персонажей и стили, а также под нереалистичные иллюстрации; реализм ей сознательно не завезли. Обучена на нескольких миллионах аниме-изображений и около 800k non-anime artistic images, без synthetic data.

Ключевые особенности

  • Аниме, иллюстрации и арт; для фотореализма не предназначена.
  • Три версии: Anima-Base для гибкости и LoRA, Anima-Aesthetic для качества по умолчанию, Anima-Turbo для быстрой генерации.
  • Нативная поддержка ComfyUI: отдельные файлы для diffusion_models, text_encoders и vae.
  • Понимает Danbooru-теги, natural language и их смесь;艺术家-теги через @, quality-теги и weighting работают.
  • LoRA лучше обучать на Base; LLM adapter тренировать не советуют, learning rate для rank 32 — около 2e-5.
  • Слабое место — длинный текст на изображении; короткие промпты могут давать нежелательный результат.

Пример

Промпт в стиле карточки: masterpiece, best quality, score_7, safe, 1girl, solo, long hair, smile, looking at viewer, simple background. Для Turbo используйте CFG 1 и 8–12 steps, для Base/Aesthetic — 30–50 steps и CFG 4–5.

ПодробнееДанные: несколько миллионов аниме-изображений и ~800k non-anime artistic images; knowledge cut-off для аниме — сентябрь 2025.
Запуск: anima-base-v1.0.safetensors → ComfyUI/models/diffusion_models, qwen_3_06b_base.safetensors → text_encoders, qwen_image_vae.safetensors → vae.
Генерация: 512²–1536² пикселей; 30–50 steps и CFG 4–5 для обычных версий.
Лицензия: CircleStone Labs Non-Commercial; изображения можно использовать коммерчески, но платный хостинг модели или API требует отдельной лицензии.
Примеры
Discuss Telegram 1 10

Will it run?

Working out the memory needs

The bot is still collecting this model's weight sizes; the estimate will appear here in a few minutes. Meanwhile you can rent a GPU with headroom.

Massed ComputeOpen Massed Compute

Run it on Massed Compute

The script downloads the model to the rented machine. How to run it is in the model card.

  1. 1. Rent a machine with a fitting GPU

    Open Massed Compute
  2. 2. Paste the script into the machine's terminal

    Connect over SSH (the Massed Compute panel shows the command) or open a terminal on the machine's desktop.

    HF_TOKEN / CIVITAI_TOKENThe script starts with token lines: paste yours between the quotes. The Hugging Face token is for gated models, the Civitai one for Civitai models.

    # jufeihux/Anima
    mkdir -p ~/models
    curl -LsSf https://astral.sh/uv/install.sh | sh
    source "$HOME/.local/bin/env"
    uv tool install "huggingface_hub[cli]"
    hf download jufeihux/Anima --local-dir ~/models/Anima
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to Anima
  3. 3. Use it

    The model is in ~/models. How to run it is in the model card (link above).

Stop the machine in the panel when you are done: it bills by the minute while it runs.

How to speed it up

Picked for this model type. Choose your hardware above to see what fits you exactly.

  • FlashAttention 44.0.0b40from November 18, 2026new release

    fastest exact attention on B200

    Kernels for Blackwell (beta). Exact attention without quantisation — fine for everything, LLMs included.

    How to enable: pip install flash-attn-4; vLLM and SGLang pick it up themselves.

    Earlier generations (2)
    • FlashAttention 3 — The Hopper (H100/H200) version with FP8.
    • FlashAttention 2 — The standard for RTX 30/40/50 and A100: exact attention without the memory cost of long contexts.
  • torch.compile2.15.0from October 28, 2026new release

    10–30% for free

    Compiles the model for your GPU. Slower first run, faster afterwards.

    How to enable: model = torch.compile(model); in ComfyUI the TorchCompileModel node.

Model family

New posts about versions of Anima (quants, finetunes, LoRAs) will show up under the bell.

No known relatives of this model yet. All derivatives on Hugging Face

About the model

jufeihux/Anima
Author
jufeihux
Category
Other
Released
Posted

On NeyroLearn

Hugging Face

101

Ask AI about this model

Answered by Gemini 4 · the chat is kept only in this tab

Comments