Beam ships first model: 501B-A23B text MoE openly released under Apache 2.0
brandondamos · x · 2026-10-06
Beam announced its first model: a competitive text-only 501B-parameter MoE with 23B active parameters, released under the Apache 2.0 license along with documentation and the full stack for running, evaluating, and fine-tuning.
The team spent a year scaling a production RL/OPD stack, running a stable training job on 10K GB300 GPUs that produced over 100M rollouts across 1M tasks — significantly improving final model performance. More details are promised in the announcement.
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