Reflection AI open-sources Beam: 501B params, 23B active, built for reasoning efficiency
SergioPaniego · x · 2026-10-06
Reflection AI has introduced Beam, a highly efficient agentic open model with 501B total parameters and only 23B active, trained end-to-end from scratch. The company says it delivers frontier reasoning efficiency and advances the Western open frontier on coding and agentic tasks, with full weights releasing this month. Commenters in open source note that Reflection AI engineers have been quietly committing to projects like transformers, vLLM, SGLang, TRL and OpenEnv for months — a "low on tweets, high on commits" lab identity behind the release.
More from coding & agent
- Dev makes every AI agent talk like Žižek via AGENT.md — generativist · 2026-11-20
- YC-backed Antinuous launches: 100x more experiments, 1,000x cheaper science — ycombinator · 2026-11-20
- Nous Research partners with Microsoft to bring Hermes Agent into Teams and Azure — NousResearch · 2026-11-20
- We Listed Our MCP Server on 53 Directories: Only 16 Went Live in 11 Days — B3N0U · 2026-11-20
- Replit Missions crawls 1,000+ e-sign flows with browser use, rebuilds them 1:1 — amasad · 2026-11-20
- Replit clones a leading AI assistant in 3 days with Missions, open-sources self-hosted agent Amuse — amasad · 2026-11-20