Skip to content
3D

LibreImageDream

из одного фото в 3D через мультивью-диффузию

1 min readupdated

LibreImageDream — перепакованная версия ImageDream (Peng Wang и Yichun Shi) в формате diffusers для LibreYOLO. По одному изображению объекта модель сэмплирует четыре вида вокруг него, а семейство LibreLGM превращает их в 3D Gaussian splats. В основе — multi-view UNet на 1.3B параметров с resampler, VAE и текстовый энкодер Stable Diffusion 2.1, а также image encoder LAION CLIP-ViT-H-14.

Ключевые особенности

  • Это зеркало: файлы byte-identical upstream, без переобучения и конвертации.
  • Из одного изображения модель строит четыре ракурса для 3D-генерации.
  • Multi-view UNet с image-prompt веткой: 1.3B параметров с resampler.
  • В комплекте компоненты SD 2.1: VAE, OpenCLIP ViT-H/14 и DDIM scheduler.
  • Лицензия CreativeML OpenRAIL++-M с use restrictions.

Пример

from libreyolo import LibreGen

model = LibreGen("lgm")          # downloads this repo and LibreLGM0.4b
model("chair.jpg").save("chair.ply")
ПодробнееМодель не новая: это mirror upstream, без retraining и изменений весов.
Для работы нужен LibreLGM0.4b — вместе они дают 3D Gaussian splats.
Лицензия CreativeML OpenRAIL++-M запрещает незаконный контент, вред несовершеннолетним, диффамацию, харассмент, дискриминацию и медицинские советы.
Бенчмарков и требований к VRAM в карточке нет.
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.

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

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

About the model

LibreYOLO/LibreImageDream
Author
LibreYOLO
Category
3D
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