EN | 中文

🔥Please check out our newly released SenseNova-SI-1.2-InternVL3-8B, achieving state-of-the-art performance among open-source models of comparable size across eight recent spatial intelligence benchmarks: VSI, MMSI, MindCube, ViewSpatial, SITE, BLINK, 3DSRBench, EmbSpatial-Bench.

Despite remarkable progress, multimodal foundation models still exhibit surprising deficiencies in spatial intelligence. In this work, we explore scaling up multimodal foundation models to cultivate spatial intelligence within the SenseNova-SI family, built upon established multimodal foundations including visual understanding models (i.e., Qwen3-VL and InternVL3) and unified understanding and generation models (i.e., Bagel). We take a principled approach to constructing high-performing and robust spatial intelligence by systematically curating SenseNova-SI-8M: eight million diverse data samples under a rigorous taxonomy of spatial capabilities. SenseNova-SI demonstrates unprecedented performance across a broad range of spatial intelligence benchmarks: 68.7% on VSI-Bench, 43.3% on MMSI, 85.6% on MindCube, 54.6% on ViewSpatial, and 50.1% on SITE, while maintaining strong general multimodal understanding (e.g., 84.9% on MMBench-En). More importantly, we analyze the impact of data scaling, discuss early signs of emergent generalization capabilities enabled by diverse data training, analyze the risk of overfitting and language shortcuts, present a preliminary study on spatial chain-of-thought reasoning, and validate the potential downstream application. SenseNova-SI is an ongoing project, and this report will be updated continuously. All newly trained multimodal foundation models are publicly released to facilitate further research in this direction. In the future, SenseNova-SI will be integrated with larger-scale in-house models.

To facilitate the research in this area, as a first step, we have released a highly effective subset, SenseNova-SI-800K. This subset captures a substantial portion of the performance gains of the full SenseNova-SI dataset while providing a manageable data scale for experimentation.

Building on this, we now release SenseNova-SI-1.1-InternVL3-8B-800K, a model trained exclusively on the SenseNova-SI-800K subset. This model is provided as a reference for researchers working with the 800K-scale dataset, enabling experiments and validation of scaling behaviors observed in SenseNova-SI.

Models trained on this subset demonstrate notable improvements over the base model and achieve competitive performance against strong spatial intelligence baselines. However, we emphasize that SenseNova-SI-1.1-InternVL3-8B (800K) is released solely for research and scaling-law analysis purposes. It is not the primary recommended model of the SenseNova-SI series, nor does it represent the full performance achievable with the complete dataset.

We recommend using uv to manage the environment.

A simple image-free test to verify environment setup and download the model.

To reproduce the benchmark results above, please refer to EASI to evaluate SenseNova-SI on mainstream spatial intelligence benchmarks.