Reflection AI Releases Beam, a 501B Open-Weight MoE Model
Nvidia-backed startup Reflection AI has released Beam, its first open-weight AI model designed for coding, reasoning, and autonomous agent tasks. Built on a sparse Mixture-of-Experts (MoE) architecture, Beam features 501 billion total parameters while activating only 23 billion parameters per token. As Western technology companies seek to counter high-performing, lower-cost open-weight models from Chinese AI labs, Beam offers developers a compute-efficient alternative tailored for software automation. Its release highlights the industry's accelerating shift toward sparse MoE designs to cut inference costs without sacrificing top-tier performance. Reflection AI claims Beam matches Zhipu AI's GLM-5.2 and approaches Qwen3.8-Max in software agent and coding benchmarks. Founded in 2024 by former DeepMind researchers Misha Laskin and Ioannis Antonoglou, the startup also secured additional compute access through SpaceX's Colossus 2 data center.
## BACKGROUND
Open-weight AI models publish their internal neural network weights, allowing developers to host and customize the software locally. Mixture-of-Experts (MoE) is a model design that routes inputs to specialized sub-networks ('experts'), keeping active parameter counts low during inference while maintaining a massive total parameter capacity.