TLDR Tech / Launches
Introducing Beam: Reflection's 501B open-weight model — Reflection
Reflection has introduced Beam, its first open-weight sparse Mixture-of-Experts model featuring 501 billion total parameters and 23 billion active parameters. Built for coding, reasoning, and agentic workloads, the model was pretrained on 23.8 trillion tokens and trained using a high-compute reinforcement learning run on 10.5K NVIDIA GB300 GPUs. The company plans to release the model weights under an Apache 2.0 license later this month.
Read the full story