ИИ 3D
Matrix-3D
Generate large-scale explorable 3D scenes with high-quality panorama videos from a single image or text prompt.
★ 838 звёздMIT
Смотреть руководство →Shap-E — проект с открытым исходным кодом из категории «ИИ 3D» от openai (openai/shap-e на GitHub). У него 12 261 звезда на GitHub и 1 079 форков, он распространяется по лицензии MIT и последний раз обновлялся на GitHub 22 июня 2024.
Shap-E — OpenAI's text-to-3D and image-to-3D model: generate 3D objects as implicit functions and render them as NeRFs or textured meshes, MIT
Clone the repo and install it in editable mode, per the Usage section: pip install -e .
Open sample_text_to_3d.ipynb for text-conditioned generation or sample_image_to_3d.ipynb for image-conditioned generation, and run the notebook.
Render the generated implicit function as a NeRF or extract a textured mesh for use in your 3D tool of choice.
One model family generates 3D objects conditioned on either a prompt or a picture.
Objects are generated as implicit functions, renderable as NeRFs or convertible to textured meshes — not locked to one format.
The encode_model notebook demonstrates encoding existing 3D models and multiview renders into Shap-E latents.
Official OpenAI code and weights for the paper, MIT licensed, with samples documented in samples.md.
The README usage route.
git clone https://github.com/openai/shap-e.git
cd shap-e
pip install -e .The example notebook for prompt-conditioned generation.
shap_e/examples/sample_text_to_3d.ipynbThe example notebook for image-conditioned generation.
shap_e/examples/sample_image_to_3d.ipynbOpenAI — this is the official code and model release for the Shap-E paper (openai/shap-e on GitHub).
Yes — MIT licensed, code and model release included.
Treat them as strong first drafts: generated objects usually need cleanup, retopology or UV work in a 3D editor before production use, like most text-to-3D output.
Against Hunyuan3D or Tripo-style newer systems, Shap-E is older and lower-fidelity but simpler, fully open (MIT) and unusually clean to study and hack. Against Point-E, its predecessor from the same lab, Shap-E produces implicit functions that render as NeRFs or meshes instead of bare point clouds.
Clone the repo, pip install -e ., then run sample_text_to_3d.ipynb with your own prompt.
Generate large-scale explorable 3D scenes with high-quality panorama videos from a single image or text prompt.
Text-to-3D that outputs editable parametric programs, not just meshes.
ICCV 2023 research: ground text descriptions to multiple 3D objects.
Give your AI agent CAD superpowers - generate real STEP/STL models from text
Tencent Hunyuan3D-1.0: A Unified Framework for Text-to-3D and Image-to-3D Generation
Text-to-CAD: generate NURBS models from plain language.
Другие проекты из категории «ИИ 3D» с наибольшим числом звёзд в каталоге LibreAI — реалистичный список альтернатив, если Shap-E вам не подходит.
Control Blender with natural language: an MCP plugin that lets any LLM model, light and animate in 3D
Pascal Editor: an open-source 3D architectural editor with MCP tools for humans and AI agents
Give your AI agent CAD superpowers - generate real STEP/STL models from text
Turn a reference image into a procedural, animation-ready Three.js model - as code
High-Resolution 3D Assets Generation with Large Scale Hunyuan3D Diffusion Models.
Соседние проекты из категории «ИИ 3D» в каталоге, ближайшие по числу звёзд к Shap-E — продолжайте изучать полку по обе стороны от него.
Official repo for paper "Structured 3D Latents for Scalable and Versatile 3D Generation" (CVPR'25 Spotlight).
Nerfstudio — the plug-and-play studio for NeRFs and Gaussian splatting: turn photos of a real scene into explorable 3D, with a modular API and viewer, Apache-2.0
High-Resolution 3D Assets Generation with Large Scale Hunyuan3D Diffusion Models.
Modly — a desktop app that generates 3D models from images or prompts, running entirely on your own GPU.
Turn a reference image into a procedural, animation-ready Three.js model - as code
From Images to High-Fidelity 3D Assets with Production-Ready PBR Material
Бесплатно и открыто — забирайте исходный код, ставьте звезду репозиторию и начинайте создавать.