Reflection AI debuts Beam model to compete with Chinese open-weight systems

Brooklyn-based startup Reflection AI unveiled its first open-weight artificial intelligence model, named Beam, on Monday. TechCrunch AI and The Information reported that the two-year-old company is positioning the system to rival open models from Chinese developers, including DeepSeek, Qwen, and Z.ai. [1][2]

TechCrunch AI reported that Reflection AI claims Beam matches the performance of leading Chinese open models on advanced reasoning benchmarks while operating at significantly lower compute costs. The Information and Finimize both noted that Reflection AI is backed by Nvidia, though The Information reported that the startup is still in the process of evaluating and testing the model. [1][2][7]

The company is targeting enterprises and sovereign nations with a product pitch centered on AI factories, according to TechCrunch AI. This initiative is designed to allow institutions to construct customized, local AI systems by training Reflection models on their own proprietary data. [1]

Financial Times, Semafor, and Fortune all reported that the release represents a broader Western push to compete with free open models originating from Chinese AI laboratories. [3][5][6]