OmniGen Family: Unified Multimodal Generation by VectorSpace Lab

OmniGen is a unified multimodal generation model series by VectorSpace Lab supporting T2I, image editing, and in-context generation.

O

OmniGen by VectorSpace Lab

Text-to-ImageImage EditingIn-Context GenerationOpen Source

Unified multimodal generation model series. OmniGen2 combines text-to-image, instruction-based image editing, in-context generation, and visual understanding in a single model with decoupled decoding pathways.

Hugging FaceGitHubPaper

Models

ModelDescriptionLicenseReleased
OmniGen2Unified multimodal gen with T2I, editing, in-contextApache 2.02025-06

Select a model above for curated weight downloads, tutorials, and related information.