Reflection AI's Beam-1.0-501B-A23B debuts at #6 among US labs on Agent Arena
arena · x · 2026-10-17
Agent Arena announces that Reflection AI's Beam-1.0-501B-A23B (501B total params, 23B active) has landed on the leaderboard, ranking #6 among US labs and #13 overall — placing Reflection AI in the top 15 labs. The release is also positioned as a contribution to the open-source ecosystem.
More from Models
- Dev asks why non-tech users prefer ChatGPT when Claude is 'far better' — prasenx · 2026-10-17
- ChatGPT's annoying regressions: garbled input, broken inline images, constant timeouts — ChristianKl · 2026-10-17
- Model tweaks barely move MMLU but easily break generative tasks — aryaman2020 · 2026-10-17
- Claude reverse-engineers 1977 DES test vectors to a PDP-11 rand() seed — chrisrohlf · 2026-10-17
- METR reports on OpenAI and Anthropic models required lab review and approval — kevinnbass · 2026-10-17
- Claude cracks 30-year mystery: ECDSA curve seeds were just RNG output — scaling01 · 2026-10-17