NVIDIA Launches Advanced AI Models and Tools for Robotics

NVIDIA has expanded its artificial intelligence suite with the launch of Nemotron 3.5 Super VL, a 120-billion-parameter open multimodal model built for enterprise agents. Designed for speed and precision, the model marks a notable step forward in how AI handles complex workflows. In tandem with the Super VL, NVIDIA also released a series of smaller models called Bolt, which are optimized to run locally on RTX systems and Jetson devices. Together, these tools are set to broaden the operational reach of AI agents in various industries.

Developed alongside the Nemotron Coalition, the Nemotron 3.5 Super VL focuses on high-level accuracy for tasks involving visual understanding and instruction following. It manages this performance at roughly one-fifth the parameter count of NVIDIA’s largest reasoning model, combining efficiency with significant processing power. The model is especially effective for high-volume work, allowing enterprise agents to finish tasks in hours that might otherwise have taken days or weeks to complete.

To complement the Super VL, the new Bolt models provide specialized solutions for local workloads. The Bolt 74B-A9B is intended for general daily tasks, balancing intelligence with quick response times, while the Bolt E4B is built specifically for conversational agents and physical hardware. These releases give organizations the ability to run AI agents across different types of hardware without losing performance, offering more flexibility in how they deploy their technology.

Alongside these models, NVIDIA introduced Flora, a tool that automates the labeling of video used to train robots. By streamlining the data preparation stage, Flora can handle up to 200,000 hours of video, helping organizations train robotic systems with greater efficiency. This combination of advanced models and support tools places NVIDIA at the center of AI development, serving a range of needs from corporate software to physical robotics.

As different sectors begin to adopt and customize Nemotron 3.5 Super VL, the scope for specialized AI applications is widening. Companies have already started using the model for tasks in cybersecurity, financial analysis, and medical claims management. Through these ongoing updates, NVIDIA continues to develop the infrastructure that allows organizations to integrate artificial intelligence more deeply into their daily operations.