SmolRGPT: Efficient Spatial Reasoning for Warehouse Environments with 600M Parameters
This work addresses efficiency and spatial understanding for warehouse and robotics applications, representing an incremental improvement with a novel method for a known bottleneck.
The paper tackles the problem of deploying vision-language models in resource-constrained environments like warehouses by introducing SmolRGPT, a compact 600M-parameter model that achieves competitive performance on spatial reasoning benchmarks, matching or exceeding larger alternatives.
Recent advances in vision-language models (VLMs) have enabled powerful multimodal reasoning, but state-of-the-art approaches typically rely on extremely large models with prohibitive computational and memory requirements. This makes their deployment challenging in resource-constrained environments such as warehouses, robotics, and industrial applications, where both efficiency and robust spatial understanding are critical. In this work, we present SmolRGPT, a compact vision-language architecture that explicitly incorporates region-level spatial reasoning by integrating both RGB and depth cues. SmolRGPT employs a three-stage curriculum that progressively align visual and language features, enables spatial relationship understanding, and adapts to task-specific datasets. We demonstrate that with only 600M parameters, SmolRGPT achieves competitive results on challenging warehouse spatial reasoning benchmarks, matching or exceeding the performance of much larger alternatives. These findings highlight the potential for efficient, deployable multimodal intelligence in real-world settings without sacrificing core spatial reasoning capabilities. The code of the experimentation will be available at: https://github.com/abtraore/SmolRGPT