Skip to main content
ModelComputer Visionvv1

Point-E

by OpenAI · open-source · Last verified 2026-03-17

Point-E is OpenAI's pioneering text-to-3D model that generates 3D point clouds from text prompts in seconds by first generating a 2D image and then lifting it to 3D, establishing a fast baseline for open-source 3D generation. While its geometric quality is limited compared to newer systems, its speed and simplicity made it a widely used research reference and starting point for subsequent 3D generation work.

https://github.com/openai/point-e
C+
C+Average
Adoption: CQuality: BFreshness: C+Citations: B+Engagement: F

Specifications

License
MIT
Pricing
open-source
Capabilities
text-to-3d, image-to-3d, point-cloud-generation, fast-inference, open-source
Integrations
huggingface, pytorch
Use Cases
3d-prototyping, game-asset-research, research, creative-exploration, rapid-ideation
API Available
Yes
Parameters
~40M
Context Window
N/A
Modalities
text, image, 3d
Training Cutoff
2022
Tags
3d-generation, text-to-3d, point-cloud, openai, open-source
Added
2026-03-17
Completeness
100%

Index Score

50.4
Adoption
48
Quality
62
Freshness
55
Citations
75
Engagement
0

Explore the full AI ecosystem on Agents as a Service