NVIDIA

NVIDIA FLARE Auto-FL is an automated, AI-driven research loop designed to test and optimize federated learning strategies.  The idea is straightforward: start with a comparable benchmark task, give the agent a clear research control plane, set a fixed training budget, constrain the mutation surface, and record every result in an experiment ledger. From there, the agent can autonomously iterate through candidate FL strategies while preserving the FLARE Client API and Recipe API contracts. Rather than handing an agent an open-ended research problem, Auto-FL begins with a fair, comparable benchma

The DGX Spark manageability framework delivers a modular stack, designed to integrate into the tools enterprise IT teams already use rather than replace them. NVIDIA partners that currently support DGX Spark from an enterprise manageability perspective include Progress Chef, Perforce Puppet, and Canonical Landscape.  The operating model is intentionally simple: agentless SSH execution with bounded standard JSON output. A resident management agent is not required to run on the DGX Spark endpoint. Instead, IT teams invoke tools over SSH, and each tool returns a standardized JSON envelope that in

Beyond the default CIFAR-10 simulation, the Auto-FL pattern is highly adaptable. By decoupling the primary control plane from the task profile—which specifies the dataset, metrics, and mutation constraints—researchers can apply the same autonomous experiment discipline to various model families without rebuilding the underlying harness. To demonstrate this flexibility, a medical visual language model (VLM) task is included in this example. This example integrates a federated Qwen3-VL LoRA training workflow into the NVIDIA FLARE client and recipe APIs. The setup simulates three distinct medical