VectorSpaceLabImage model
OmniGen2
Arena ratingnot rated
n/aUser voteshigher is better
GraduallyBenchnot tested yet
n/aour own rating0 to 100
Priceno price
n/aper imagelower is better
Summary
OmniGen2 is an image model from VectorSpaceLab. The T2I example supports a negative prompt and recommends a default value for it.
On this page
Head-to-head comparisons
Show all comparisons (139)
All specifications
Licence and access
Further details
- Generation
- The official project documents a prompt-only text-to-image mode alongside multimodal and image-conditioned generation.
- Source: Generation
- Web demo
- The project links a web application and Gradio demos; no API model ID is documented. Current runtime availability was not checked.
- Source: Web demo
- Negative prompt
- The T2I example supports a negative prompt and recommends a default value. Set it explicitly empty for unchanged benchmark prompts.
- Source: Negative prompt
Sources and data date
| Type | Evidence and data date |
|---|---|
| Additional source | VectorSpaceLab (retrieved November 1, 2026) |
| Additional source | VectorSpaceLab (retrieved November 1, 2026) · VectorSpaceLab · Editorial description reviewed November 1, 2026 |