SenseTime has officially introduced its SenseNova-U1 Pro model, heralded as the industry’s pioneering multimodal intelligent agent foundation model, featuring an inherent unified core of "Understanding-Generation-Action." The model is slated to commence invitation-based testing in July 2026. Drawing parallels to GPT-Image 2, this model underscores a "delivery-grade" design philosophy, boasts native support for 8K resolution output, and facilitates an extended cycle of design, generation, and review, effectively operating as a "thinking designer."
