Skip to content
Other

MOSS-Image-SFT-lora-Anime

аниме-стиль для MOSS-Image-SFT в ComfyUI

1 min read

LoRA-адаптер для базовой модели MOSS-Image-SFT, сдвигающий генерацию в аниме / 2D-стиль. Обучен на 2000 отобранных вручную аниме-иллюстрациях за 10 000 шагов, формат safetensors, целевая среда — ComfyUI.

Ключевые особенности

  • 2000 аниме / 2D-иллюстраций в датасете, 10 000 шагов обучения.
  • Чекпоинт каждые 1000 шагов — всего 10 файлов с разной силой стиля.
  • Рекомендуемый старт — чекпоинты 6000–8000 шагов: стабильный стиль и баланс детализации с обобщением.
  • Ранние чекпоинты (1000–4000) дают лёгкую стилизацию, поздние (9000–10000) — максимальную, но с риском переобучения.
  • Триггер-слово <trigger> нужно заменить на реальное; если оно не задавалось, стиль регулируется весом LoRA.

Пример

Промпт из карточки:

<trigger>, 1girl, solo, short brown hair, amber eyes, hoodie, sitting, bedroom, warm lighting, soft smile, anime style, clean lineart, detailed eyes, masterpiece, best quality

Негатив: low quality, blurry, bad anatomy, extra fingers, watermark, text

ПодробнееВес LoRA: strength_model и strength_clip по 0.7–0.9, по умолчанию 0.8 / 0.8.
Шаги, CFG и разрешение — по рекомендациям базовой MOSS-Image-SFT.
Ограничения: фотореализм, 3D и «толстая живопись» поддерживаются слабо; конкретные персонажи и IP не гарантируются.
Лицензия: личное обучение и некоммерческое использование, совместимость с лицензией базовой модели нужно проверять самостоятельно.
Discuss Telegram 0

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.

    # mfxq2l/MOSS-Image-SFT-lora-Anime
    # ===== Tokens: paste yours between the quotes =====
    export HF_TOKEN=""        # huggingface.co/settings/tokens, for gated models
    # ==========================================
    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 mfxq2l/MOSS-Image-SFT-lora-Anime --local-dir ~/models/MOSS-Image-SFT-lora-Anime
    # The model is in ~/models. How to run it is in the model card (link above).
  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 MOSS-Image-SFT-lora-Anime (quants, finetunes, LoRAs) will show up under the bell.

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

About the model

mfxq2l/MOSS-Image-SFT-lora-Anime
Author
mfxq2l
Category
Other
Released
Posted

On NeyroLearn

Hugging Face

00

Ask AI about this model

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

Comments

Images

JANK2

JANK2