FULL STORY

SenseNova U1 Pro Debuts with Open-Source Visual Dataset

SenseNova launched the U1 Pro multimodal base model at WAIC 2026 and open-sourced a 50M-sample visual dataset to advance complex multimodal creation.

2026-07-18 ~ 2026-07-21 · 2 episodes · 12 posts

Episode 1 · SenseTime Launches SenseNova U1 Pro Native 8K Multimodal Model (2026-07-18, 6 posts)

SenseTime unveiled its flagship image generation model, SenseNova U1 Pro, at WAIC. Designed for complex multimodal creation and delivery, the model has garnered attention for integrating native 8K output, complex layout processing, and Chinese long-text typography into a single foundational architecture that covers the entire workflow from understanding and planning to generation and correction.

Key Capabilities and Details

According to related posts, U1 Pro supports native 8K resolution output and ultra-wide aspect ratios, targeting high-resolution, complex composition tasks like posters, infographics, and long scrolls. SenseTime emphasized its professional design aesthetics and precise handling of complex layouts. Deliprao highlighted its agentic image generation for long tasks and iterative refinement through "interleaved thinking and generation."

Positioning and Demonstrations

Xinzhiyuan noted that SenseTime positions U1 Pro as a next-generation multimodal model unifying "seeing, generating, and acting" in one base, demonstrating its capabilities in 8K scrolls, movie posters, infographics, and character design. Qubit described it as a "delivery system" for complex multimodal tasks, handling understanding, planning, information organization, generation, and final correction.

Benchmarks and Comparisons

In the image generation domain, U1 Pro is compared against GPT Image 2 and Seed Dream. Deliprao relayed official claims that it can compete with these top-tier models, and teortaxesTex mentioned an official blog stating its results are comparable to GPT Image 2. These comparisons currently stem from official presentations and secondary reports, lacking independent evaluation results in the provided posts.

Episode 2 · SenseTime Launches Multimodal Agent Base U1 Pro and Open-Source Vision Dataset (2026-07-20, 6 posts)

During WAIC 2026, SenseTime launched the multimodal agent base model "SenseNova U1 Pro" and open-sourced SenseNova-Vision, a visual dataset containing 50 million samples. The model aims to elevate multimodal systems from simply "providing answers" to directly "delivering finished products," drawing significant industry attention.

Core Technology and Product Positioning

Built on the NEO-Unity architecture, U1 Pro unifies understanding, generation, and action into a native interleaved image-text reasoning capability. Unlike conventional entertainment-grade image generation, this model focuses on native 8K output and "delivery-grade" high-precision visual content generation. It is primarily designed for long-horizon delivery tasks, capable of handling complex Chinese text, infographics, commercial posters, academic layouts, and industrial diagrams. Compared to the previous U1 Lite version, officials specifically highlighted its enhanced control over image-text details.

Practical Applications and Review Feedback

According to hands-on tests and reports from media outlets like APPSO, U1 Pro demonstrated the ability to generate complex visual content such as World Cup match reports and original movie posters. Tech publication Xin Zhiyuan noted that this marks SenseTime's multimodal technology achieving practical potential in complex tasks—transitioning directly from concept comprehension to the delivery of a final visual product.