NVIDIA Launches NeMo Switchyard for AI Workload Routing
NVIDIA has released NeMo Switchyard, an open-source library designed to optimize AI agent workloads by routing requests based on model capability, latency, and cost. This initiative aims to enhance the efficiency of AI applications by automatically selecting the best models for specific tasks. Alongside this, NVIDIA has introduced the Nemotron 3.5 Lightning model, which supports high-volume agentic AI workflows.
NVIDIA released the open-source library NeMo Switchyard to help make AI agent applications more efficient. It works by routing AI workloads to different models based on their capability, latency, and cost. As AI evolves from simple chatbots into more complex autonomous agents, managing models effectively has become a greater challenge. NeMo Switchyard gives developers a way to build custom routing systems that send each request to the most suitable AI model, which helps optimize performance and allocate resources.
The release of NeMo Switchyard came as NVIDIA expanded its Nemotron model family with Nemotron 3.5 Lightning. This model has 30 billion parameters and is designed for efficient, long-running agentic workflows. Its development is part of NVIDIA's work on open models, aiming for greater accuracy and speed on specialized tasks inside multi-agent systems. Nemotron 3.5 Lightning can perform targeted functions like code review and security monitoring, making it useful for enterprises that want to expand their AI capabilities.
The NeMo Switchyard library has drawn interest from a number of industries. Companies are using it to streamline their AI workflows by directing tasks to the most capable model available, whether it is open-source, proprietary, or one of NVIDIA's. Because developers do not need to rewrite their applications to use it, NeMo Switchyard simplifies integration and allows for faster deployment of AI solutions.
Used together, NeMo Switchyard and Nemotron 3.5 Lightning can improve the speed and accuracy of agentic tasks.
Enjoyed this story?
Show the newsroom a little love — one tap per reader.
Liam covers climate solutions and the people helping the planet heal.
Be part of the good
Stories like this start with people who care. Share it, or submit your own uplifting story to inspire millions today.
