Reflection AI Beam model announcement

Reflection AI introduced Beam on October 5 as its first open-weight model. The company describes it as a sparse mixture-of-experts system with 501 billion total parameters and 23 billion active parameters, built for coding, reasoning and agentic tasks. Beam is still undergoing final red-teaming and evaluation. Reflection is accepting early-access requests and says it plans to publish the weights, technical report, model card and developer materials later in October. The announcement provides company-run benchmark comparisons, but the weights and full evaluation materials are not yet available for independent testing.