From the source
Reflection AI introduced Beam, an open-weight sparse Mixture-of-Experts model with 501 billion total parameters and 23 billion active parameters, focused on coding, reasoning, and agentic tasks.
The model was pretrained on 23.8 trillion tokens and trained with high-compute reinforcement learning using 10.5K NVIDIA GB300 GPUs over four weeks, generating over 100 million rollouts.
Beam is undergoing final red-teaming and evaluations, with weights, technical report, and developer artifacts to be released later this month.