OpenEMMA introduces a resource-efficient approach to build end-to-end autonomous driving systems using multimodal large language models (MLLMs).

• Enables processing of visual data and reasoning about complex driving scenarios
• Implements novel parameter-efficient fine-tuning methods that require fewer resources
• Delivers an open-source solution that democratizes autonomous driving research
• Creates a new paradigm for end-to-end autonomous driving systems

Engineering Impact: OpenEMMA represents a significant advancement in autonomous vehicle technology by combining visual processing with reasoning capabilities in a resource-efficient package, potentially accelerating the development and deployment of self-driving cars.

OpenEMMA: Open-Source Multimodal Model for End-to-End Autonomous Driving