Reflection AI releases Beam: 501B-param open-weight model under Apache 2.0, trained on 23.8T tokens
shakoistsLog · x · 2026-10-06
Reflection AI has released Beam, an open-weight agentic model with 501B total / 23B active parameters under Apache 2.0, trained end-to-end from scratch and claiming frontier reasoning efficiency plus Western open-source leadership on coding and agentic tasks. Full weights ship this month. Team member nayshins detailed the data engineering: 23.8T training tokens, an in-house OCR pipeline generating trillions of tokens from hundreds of millions of PDFs, and a data platform scaled from single terabytes to multiple petabytes, all built from zero in 8 months.
More from Infra
- Nvidia down ~40% from October peak as Anthropic IPO pops 34% above expectations — The Verge AI · 2026-11-20
- Ion Stoica explains why complex AI workloads increasingly fit Ray's design — PyTorch · 2026-11-20
- GPU rents up, token prices down: inference providers' break-even load jumps 22%–75% in 5 months — AccBalanced · 2026-11-20
- Missing config file was throttling GPUDirect Storage — fix yields 3x reads, 9x writes — AccBalanced · 2026-11-20
- NVIDIA Dynamo cuts RL rollout bottlenecks with KV-aware routing and delta weight updates — NVIDIAAI · 2026-11-20
- Arm reaches the AI laptop: Surface Laptop Ultra runs 120B models locally — kimmonismus · 2026-11-20