Reflection AI releases Beam: 501B-total/23B-active open weights under Apache 2.0
zacharynado · x · 2026-10-17
Reflection AI has shipped Beam, its first open-weight model: 501B total parameters with 23B active (MoE), released under Apache 2.0 with the model card live on Hugging Face, targeting coding, reasoning and agentic tasks. The company calls it an incredibly token-efficient model in the Western open frontier. It also open-sourced Mirror, its new training/serving harness described as blazing fast.
Related event: Reflection AI open-sources 501B MoE model Beam-1.0 under Apache 2.0(45 posts)→
More from Infra
- Ben Horowitz says Amazon is sold out of compute through end of 2028 — RihardJarc · 2026-10-17
- Free 8-hour AI data center course launches on freeCodeCamp, from grid power to GPU economics — trawasthi_ai · 2026-10-17
- Broadcom Wi-Fi bug enables passwordless remote code execution; ~1.5M Belgian routers unpatched — jedisct1 · 2026-10-17
- Local LLM hosting isn't always cheaper: cached-token rate decides the break-even point — paq85 · 2026-10-17
- DeepSeek's new checkpoint cuts CoT token use by up to 10x, sparking Jevons paradox talk — teortaxesTex · 2026-10-17
- Latent Space: Meta's 816B Open Model Tops US Charts, TPUv7 Beats GB300 on $/perf — Latent Space · 2026-10-17