Skip to content
Other

Remi Ayasaki LoRA

персонаж из Horimiya для Illustrious SDXL

2 min read

Неофициальный character LoRA от abbuibuibui для Реми Аясаки из Horimiya. Обучен на waiIllustriousSDXL v17, в репозитории лежат веса E1–E7; по сравнительной оценке дефолт — E6 при силе 0.8. Это фанатская производная, права на персонажа остаются у правообладателей.

Ключевые особенности

  • Семь эпох E1–E7: E6 рекомендуется как основная, E7 очень близка к ней.
  • Триггер личности remi_ayasaki; outfit-токены: remi_ayasaki_winter_uniform, remi_ayasaki_winter_coat_scarf, remi_ayasaki_cooking_apron, remi_ayasaki_red_tracksuit.
  • Настройки: Euler a, 28 steps, CFG 4.5, CLIP skip 2, 832×1216; сила 0.8 для лица и зимней формы, 0.6 — если нужно белое пальто.
  • Обучение: LoRA rank 64 / alpha 32, UNet и оба text encoder, 59 изображений × repeat 3 × 7 эпох, 406 optimizer steps.
  • Совместимость: лучше всего waiIllustriousSDXL; Illustrious/NoobAI merges — 0.7–0.9; Pony V6 XL — возможен дрейф; generic SDXL/SD1.5 — плохо или несовместимо.

Пример

Промпт для E6 на силе 0.8:

<lora:remi_ayasaki-000006:0.8>, masterpiece, best quality, very aesthetic, absurdres, anime screencap, 1girl, solo, remi_ayasaki, remi_ayasaki_winter_uniform, twintails, long hair, pink hair, red eyes, looking at viewer
ПодробнееФайлы по 456 495 868 байт (~456 МБ); E6 — рекомендуемый дефолт.
Пальто с шарфом: при 0.6 — белое с красным шарфом, при 0.8 — бежевое, шарф бледнеет.
Без outfit-токена при 0.8 остаётся розовый свитер; без имени утечка лица и волос возможна уже при 0.4.
Совместимость: waiIllustriousSDXL — лучшая; Illustrious/NoobAI merges — 0.7–0.9; Pony V6 XL — возможен дрейф; generic SDXL/SD1.5 — плохо/несовместимо.
Обучение: 59 изображений, rank 64 / alpha 32, UNet и оба text encoder, 406 optimizer steps, kohya sd-scripts на DGX.
Примеры
Discuss Telegram 0 119

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.

    # abbuibuibui/RemiAyasaki-Lora
    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 abbuibuibui/RemiAyasaki-Lora --local-dir ~/models/RemiAyasaki-Lora
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to RemiAyasaki-Lora
  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 RemiAyasaki-Lora (quants, finetunes, LoRAs) will show up under the bell.

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

About the model

abbuibuibui/RemiAyasaki-Lora
Author
abbuibuibui
Category
Other
Released
Posted

On NeyroLearn

Hugging Face

1190

Ask AI about this model

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

Comments