Skip to content
3D

Forma3D Tiny 3

крошечная image-to-3D модель на 306 тыс. параметров

1 min readupdated

Продолжение Lucasgitguy-co/forma3d-tiny-2 от автора Lucasgitguy-co: та же архитектура на 306 321 параметр, сетка 16³, ширина 64 и до 4096 Gaussian slots. Модель принимает изображение и выдаёт Gaussian PLY; обучение шло 6000 шагов на 512 объектах ABO.

Ключевые особенности

  • Всего 306 321 параметр — очень компактная архитектура для image-to-3D.
  • До 4096 Gaussian slots на сетке 16³; результат — Gaussian PLY, а не текстурированный GLB.
  • Обучение на 512 объектах ABO и 2048 ортографических proxy-изображениях на чёрном фоне.
  • Нет conditioning по текстовому промпту — только изображение.
  • Ожидаемый вход: центрированный объект на чёрном фоне.
  • Рецепт Mac-only, запуск возможен на устройстве mps.

Пример

pip install .
image3d generate --checkpoint model.pt --image centered-object-on-black.png --out object.ply --device mps
image3d preview --ply object.ply --out preview.png
ПодробнееMean test loss: 0.690551 → 0.544487; view occupancy IoU при пороге 0.5: 0.286866 → 0.339860.
Вывод — Gaussian PLY, не текстурированный GLB; текстовых промптов нет.
Нужны центрированные объекты на чёрном фоне; рецепт Mac-only с устройством mps.
Метрики не подтверждают фотореализм, качество меша или реконструкцию по естественным фото.
Discuss Telegram 0 45

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.

    # Lucasgitguy-co/forma3d-tiny-3
    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 Lucasgitguy-co/forma3d-tiny-3 --local-dir ~/models/forma3d-tiny-3
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to forma3d-tiny-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 forma3d-tiny-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

Lucasgitguy-co/forma3d-tiny-3
Author
Lucasgitguy-co
Category
3D
Released
Posted

On NeyroLearn

Hugging Face

450

Ask AI about this model

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

Comments