Reflection Open-Sources 501B Model Beam: Only 23B Parameters Activated per Token, Claims to Match GLM-5.2 with a Quarter of the Compute
Reflection AI has open-sourced Beam, a 501B-parameter sparse MoE model under Apache 2.0 that activates only 23B parameters per token and targets coding and agentic workloads with a 1M-token context window. The company claims it matches the 744B GLM-5.2 on inference benchmarks using one-quarter to one-third of the compute.