Skip to content
Other

Rigless LoRA v1

замена человека в реальном видео по одной фотографии

2 min read

LoRA для MiniMax H3 от StephenShumway. Заменяет человека в реальной съёмке персонажем с картинки и сделана для проекта Rigless. Это доработанная H3 Relay Character Swap LoRA v1.5 от akatz, дополнительно обученная на реальных клипах; файл — rigless-lora-v1.safetensors.

Ключевые особенности

  • Построена на H3 Relay Character Swap LoRA v1.5 и дообучена на реальных клипах.
  • Учит H3 переносить лицо, волосы и одежду с фото, сохраняя свет и сцену реальной съёмки.
  • В Rigless LoRA скачивается и загружается автоматически для каждого рендера.
  • В слепом тесте разработчика её предпочли версии v1.5 и варианту без swap LoRA: 3 клипа, по одному дублю.
  • По идентичности лица измеримого улучшения нет.

Пример

В ComfyUI LoRA кладут в models/loras и загружают с силой 1.0 на MiniMax H3 (ref2va) вместе с нодами H3 Relay и 8-step H3 Turbo LoRA. Промпт для одной картинки:

Replace the selected person in <Video 1> with the person in <Picture 1>. Keep the reference person's face, hair and clothing. Follow the source performance's pose, gaze, expression and movement. Preserve the scene and the other people.
ПодробнееСлепой тест: 3 клипа, по одному дублю; оценка разработчика — предпочтение перед v1.5 и без swap LoRA.
Лицензия MiniMax H3 Community License: использование H3 в США, ЕС, Великобритании и Южной Корее требует отдельной авторизации MiniMax.
Нужны права на видео и фото, согласие людей; AI-видео нужно явно помечать.
Требования к VRAM/железу в карточке не указаны.
Discuss Telegram 2 56

Will it run?

Working out the memory needs

This is a LoRA: it only runs on top of the base model akatz-ai/MiniMax-H3-Character-Swap-LoRA. The base model's size is still being collected; the estimate will appear within a day.

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.

    # StephenShumway/rigless-lora
    # ===== 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 akatz-ai/MiniMax-H3-Character-Swap-LoRA --local-dir ~/models/MiniMax-H3-Character-Swap-LoRA
    hf download StephenShumway/rigless-lora --local-dir ~/models/rigless-lora
    # 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.

MiniMax H3 prompts

All prompts

Ready prompts with the resulting video: copy and try.

Model family

Requires the base model

akatz-ai/MiniMax-H3-Character-Swap-LoRA

This model is a LoRA / adapter. Download the base model first, then load this one on top.

  1. MiniMax-H3family root
  2. MiniMax-H3
  3. MiniMax-H3-Character-Swap-LoRA
  4. rigless-lora

New posts about versions of MiniMax-H3 (quants, finetunes, LoRAs) will show up under the bell.

Quantised versions 12All on Hugging Face
Fine-tunes 12All on Hugging Face
LoRAs and adapters 12All on Hugging Face
Merges 3All on Hugging Face

About the model

StephenShumway/rigless-lora
Author
StephenShumway
Category
Other
Released
Posted

On NeyroLearn

Hugging Face

562

Ask AI about this model

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

Comments