What is simscale?

SimScale is a research project that teaches self-driving car AI systems to drive better by training them on large amounts of simulated driving data alongside real-world driving data. The core problem it addresses is that real-world driving data is expensive and slow to collect, while purely simulated data often does not transfer well to real roads. SimScale bridges that gap with a co-training strategy, and the accompanying paper was accepted as an oral presentation at CVPR 2026, a major computer vision research conference. The project builds a simulation pipeline that generates diverse, realistic looking driving scenarios, complete with reactive vehicles that respond to the ego car's actions. These simulated scenes come with pseudo-expert demonstrations, meaning the simulation also provides example trajectories of how a skilled driver would handle each situation. These simulated examples are then mixed with real driving data during training, so the AI model gets the best of both worlds: variety and scale from simulation, real-world texture from actual recordings. The result is an AI driving planner that generalizes better to challenging situations it may never have seen in real data alone. The README includes a model zoo comparing several existing end-to-end planners, such as DiffusionDrive and GTRS-Dense, before and after adding simulated data to their training, evaluated on the NAVSIM v2 benchmark's harder and standard test splits. The project releases the dataset, pretrained model checkpoints, and training code so researchers can reproduce the results or fine-tune the models on their own planners. It is aimed at autonomous driving researchers who work with end-to-end driving models and want to understand how sim-to-real co-training affects performance at scale. Data and models are also mirrored on Hugging Face and ModelScope.