SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics

Papers with Code Trending 06/02/25, 04:30 PM Papers

Summary

SmolVLA is a compact vision-language-action model that achieves competitive robotic control performance at reduced computational cost, enabling deployment on consumer-grade hardware. It introduces asynchronous inference and leverages community-collected datasets.

Vision-language models (VLMs) pretrained on large-scale multimodal datasets encode rich visual and linguistic knowledge, making them a strong foundation for robotics. Rather than training robotic policies from scratch, recent approaches adapt VLMs into vision-language-action (VLA) models that enable natural language-driven perception and control. However, existing VLAs are typically massive--often with billions of parameters--leading to high training costs and limited real-world deployability. Moreover, they rely on academic and industrial datasets, overlooking the growing availability of community-collected data from affordable robotic platforms. In this work, we present SmolVLA, a small, efficient, and community-driven VLA that drastically reduces both training and inference costs, while retaining competitive performance. SmolVLA is designed to be trained on a single GPU and deployed on consumer-grade GPUs or even CPUs. To further improve responsiveness, we introduce an asynchronous inference stack decoupling perception and action prediction from action execution, allowing higher control rates with chunked action generation. Despite its compact size, SmolVLA achieves performance comparable to VLAs that are 10x larger. We evaluate SmolVLA on a range of both simulated as well as real-world robotic benchmarks and release all code, pretrained models, and training data.

Original Article

View Cached Full Text

Cached at: 07/20/26, 09:42 AM

Paper page - SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics

Source: https://huggingface.co/papers/2506.01844

Abstract

SmolVLA is a compact, efficient vision-language-action model that achieves competitive performance at reduced computational costs and can be deployed on consumer-grade hardware.

Vision-language models(VLMs) pretrained on large-scalemultimodal datasetsencode rich visual and linguistic knowledge, making them a strong foundation for robotics. Rather than trainingrobotic policiesfrom scratch, recent approaches adapt VLMs into vision-language-action (VLA) models that enablenatural language-driven perceptionand control. However, existing VLAs are typically massive--often with billions of parameters--leading to high training costs and limited real-world deployability. Moreover, they rely on academic and industrial datasets, overlooking the growing availability of community-collected data from affordable robotic platforms. In this work, we present SmolVLA, a small, efficient, and community-driven VLA that drastically reduces both training and inference costs, while retaining competitive performance. SmolVLA is designed to be trained on a single GPU and deployed on consumer-grade GPUs or even CPUs. To further improve responsiveness, we introduce anasynchronous inferencestack decoupling perception and action prediction fromaction execution, allowing higher control rates with chunked action generation. Despite its compact size, SmolVLA achieves performance comparable to VLAs that are 10x larger. We evaluate SmolVLA on a range of both simulated as well as real-world robotic benchmarks and release all code, pretrained models, and training data.

View arXiv page View PDF Project page GitHub26k Add to collection

Get this paper in your agent:

hf papers read 2506\.01844

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash