<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0"><channel><title>Reflection debuts Beam, an open-weight AI model to rival Chinese models at lower compute cost — Live Feed</title><link>https://www.live-feeds.com/feed/reflection-debuts-beam-an-open-weight-ai-model-to-rival-chinese-models-at-lower-compute-cost</link><atom:link xmlns:atom="http://www.w3.org/2005/Atom" href="https://www.live-feeds.com/feed/reflection-debuts-beam-an-open-weight-ai-model-to-rival-chinese-models-at-lower-compute-cost/rss.xml" rel="self" type="application/rss+xml"/><description>Continuously updated, source-cited coverage.</description>
<item><title>Reflection AI Unveils Beam Open-Weight Model</title><link>https://www.live-feeds.com/feed/reflection-debuts-beam-an-open-weight-ai-model-to-rival-chinese-models-at-lower-compute-cost</link><guid isPermaLink="false">https://www.live-feeds.com/feed/reflection-debuts-beam-an-open-weight-ai-model-to-rival-chinese-models-at-lower-compute-cost#u110435</guid><pubDate>Thu, 08 Oct 2026 10:45:24 +0000</pubDate><description>Startup Reflection AI has launched Beam, a 501 billion parameter open-weight artificial intelligence model designed to challenge leading Chinese open models while operating at a lower compute cost. Built as a sparse Mixture-of-Experts architecture with 23 billion active parameters, Beam targets coding, reasoning, and agentic workloads. The model was trained on 23.8 trillion tokens and utilized a high-compute reinforcement learning run generating over 100 million rollouts across 10,500 Nvidia GB300 GPUs during a four-week training period. Reflection is aiming Beam and future models at enterpris</description></item>
</channel></rss>