Evaluating Open Models with Artificial Analysis | Nemotron Labs

Independent benchmarks cut through the noise, and few do it as rigorously as Artificial Analysis. In this session, Artificial Analysis joins to share how they evaluate open models like Nemotron: what the numbers show, how models stack up against the open weights field, and what the results mean for developers building long-running agent pipelines. What you'll learn: How Artificial Analysis constructs evaluations for agentic use cases versus general intelligence How to use benchmark data to choose the right model for execution-layer versus orchestration-layer agent roles What near-lossless NVFP4 quantization means in practice for measured agentic performance Have questions about how Artificial Analysis evaluates open models, or what the numbers mean for your own model selection? Drop them live, and the team will answer in real time.