Skip to content
Video

WAN 2.2 Action LoRAs

7 адаптеров для WAN 2.2 14B

1 min read

Репозиторий zero1101001 содержит набор из 7 LoRA-адаптеров для text-to-video и image-to-video моделей WAN 2.2 14B. Это action-специализированные адаптеры для взрослого контента (18+) с вариантами high-noise и low-noise, а также один style-LoRA; общий размер — около 2.6GB.

Ключевые особенности

  • 7 LoRA: 6 action-адаптеров и 1 style-адаптер для WAN 2.2 T2V/I2V 14B.
  • Варианты high-noise для разнообразия и low-noise для более стабильного результата.
  • Размеры файлов: 293MB для rank-16 и 586MB для повышенного rank.
  • Точность BF16; заявлена совместимость с WAN 2.2 T2V и I2V 14B.
  • Промпты в карточке ориентированы на английский язык; контент только 18+.

Пример

from diffusers import DiffusionPipeline
import torch
pipe = DiffusionPipeline.from_pretrained("path/to/wan22-t2v-14b", torch_dtype=torch.bfloat16, device_map="auto")
pipe.load_lora_weights("path/to/action-lora.safetensors")
video = pipe(prompt="POV perspective, smooth camera movement, cinematic lighting", num_inference_steps=50, guidance_scale=7.5, num_frames=24).frames
ПодробнееНа RTX 4090 генерация 24 кадров 720p занимает около 28 с для LoRA 293MB и около 32 с для 586MB; VRAM — примерно 18–19GB.
Минимум: RTX 3060 12GB VRAM и 16GB RAM; рекомендуется RTX 3090/4070 Ti и 32GB RAM.
Нужны Python 3.9+, PyTorch 2.0+, diffusers 0.25.0+; базовая модель WAN 2.2 занимает 14GB в FP8 или 27GB в FP16.
Ограничения: только 18+, соблюдение локальных законов и запрет на non-consensual/deepfake-контент.
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.

    # zero1101001/wan22-fp8-t2v-loras-nsfw
    # ===== 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 zero1101001/wan22-fp8-t2v-loras-nsfw --local-dir ~/models/wan22-fp8-t2v-loras-nsfw
    # 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.

  • SageAttention 3v2.2.0from October 28, 2025

    even faster on RTX 50 / Blackwell

    FP4 attention for Blackwell. Faster than SageAttention 2, but the authors suggest v2 when accuracy matters.

    How to enable: Build from the sageattention3_blackwell folder of the repo; in ComfyUI via KJNodes.

    Earlier generations (2)
    • SageAttention 2++ — Quantised (INT8/FP8) attention with no visible quality loss. The main speed-up for image and video generation.
    • SageAttention 1 — First version: INT8 for Q·K. Still works, but 2++ is faster and more accurate.
  • 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 (3)
    • 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.
    • xFormers — The older way to speed up attention, works almost everywhere. Sage or Flash are faster where available.
  • cache-dit (DBCache)v1.5.6from November 20, 2026new release

    1.5–3× at nearly the same quality

    Reuses block outputs between diffusion steps. Supports most DiT models.

    How to enable: pip install cache-dit; one line cache_dit.enable_cache(pipe) for diffusers.

    Earlier generations (2)
    • MagCache — Skips steps by magnitude of change. More accurate than TeaCache, no calibration.
    • TeaCache — The first popular diffusion step cache. Works, newer methods are more accurate.
  • LightX2V step distill0.6.1from November 9, 2026new release

    5–10× fewer steps

    Distilled LoRAs and an engine: video in 4–8 steps instead of 30–50.

    How to enable: Load their step-distill LoRA, CFG 1.0, 4–8 steps.

  • 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 wan22-fp8-t2v-loras-nsfw (quants, finetunes, LoRAs) will show up under the bell.

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

About the model

zero1101001/wan22-fp8-t2v-loras-nsfw
Author
zero1101001
Category
Video
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