Reka Edge

Reka Edge is Reka's lightweight multimodal AI model, positioned as the most efficient option in the Reka family. As a tier-optimized model, it's designed for scenarios where speed and cost efficiency take priority over maximum capability, while still maintaining multimodal functionality across text, images, and video inputs. The model operates with a 16,384 token context window and supports three input modalities: text, image, and video processing. This multimodal capability allows it to understand and reason about visual content alongside text, making it suitable for applications that need to process mixed media efficiently. However, it does not include tool calling or function execution capabilities, keeping the architecture focused on core inference tasks. Reka Edge targets use cases where organizations need multimodal AI capabilities but with faster response times and lower computational overhead than larger models. It competes in the lightweight category against other efficient multimodal models, offering video processing capabilities that are less common in this tier of models.