Beam: Reflection's 501B open-weight model

Reflection has introduced Beam, its first open-weight model. Beam is a sparse Mixture-of-Experts model with 501 billion total parameters and 23 billion active parameters, designed for coding, reasoning, and agentic workloads. It features fully asynchronous execution, where agents generate rollouts while the trainer learns and publishes new model versions. Each token is tagged with the version that produced it, allowing the training algorithm to account for policy staleness as completed rollouts flow into training.