Scale Collaborates With NVIDIA to Bring Nemotron to Enterprises

Scale AI is collaborating with NVIDIA to help enterprises customize the new Nemotron 3.5 Super VL open models for specific business workflows. This is enabled through a new offering, Scale Post-Training, which specializes general-purpose models using expert evaluation, data, and NVIDIA's tools. The goal is to provide enterprises with more control over quality, cost, and performance by creating specialist models that run in their own environments. Early results with clients like Skydance and EY show that post-trained Nemotron models can match or outperform frontier models at a fraction of the inference cost and with faster response times.