Skip to content
Other

moss-pixelart-v.0.3

LoRA для чистого пиксель-арта с жёсткой сеткой 8×8

1 min read

moss-pixelart-v.0.3 — LoRA от finger-gun для модели OpenMOSS-Team/MOSS-Image-SFT. Она дообучена под генерацию аккуратного пиксель-арта и 2D-ассетов для игр: каждый блок 8×8 выходных пикселей соответствует одному логическому пикселю.

Ключевые особенности

  • Триггер pxl8x или [pxl8x] в начале промпта включает стиль.
  • Строгая привязка к сетке: 512×512 даёт 64×64, 768×768 — 96×96, 1024×1024 — 128×128 нативных пикселей.
  • Рекомендуемые настройки MOSS-Image: Euler, 30–50 шагов, CFG 3.0–6.0, flow shift 5.0, LoRA weight 1.0.
  • Совместима с MOSS Turbo / DMD (8-step): можно ставить 8 шагов, сохраняя Euler и flow shift 5.0.
  • Для чистки возможных «микселей» карточка советует Sprite Fusion Pixel Snapper после генерации.

Пример

Промпт из карточки: pxl8x an isometric cyberpunk ramen stall at night, glowing neon sign reading «NOODLES», steam rising from bowls, rain puddles on asphalt, crisp pixel art

ПодробнееОбучение: 3000 шагов на 30 спрайтах 64×64, апскейленных до 512×512; LoRA rank 32/32, conv 16/16; обучался только Transformer/UNet.
Генерируйте размеры, кратные 16, лучше — 512×512, 768×768 или 1024×1024.
На 512×512 логический холст 64×64: текст в изображении лучше держать коротким (1–2 слова) или генерировать в 1024×1024.
Данных о VRAM и бенчмарках в карточке нет.
Discuss Telegram 4 236

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.

    # finger-gun/moss-pixelart-v.0.3
    # ===== 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 finger-gun/moss-pixelart-v.0.3 --local-dir ~/models/moss-pixelart-v.0.3
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to moss-pixelart-v.0.3
  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-pixelart-v.0.3 (quants, finetunes, LoRAs) will show up under the bell.

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

About the model

finger-gun/moss-pixelart-v.0.3
Author
finger-gun
Category
Other
Released
Posted

On NeyroLearn

Hugging Face

2364

Ask AI about this model

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

Comments