Latent Space · 2026/10/6 06:28:43
Reflection 发布 Beam：501B 参数美国自研开源模型
AI 初创公司 Reflection 正式推出名为 Beam 的大语言模型，拥有 501B 总参数与 23B 激活参数。该模型完全在美国从零训练，旨在填补本土开源大模型的空白，尽管在综合性能上略逊于 GLM 5.3 等最新 SOTA 模型，但标志着其作为新实验室的实质性落地。
报道全文原始报道全文
It’s been over a year since Reflection launched with us with big goals on coding (and hinted about their RL approach):
But they stayed “stealth” longer than Thinking Machines and it was not clear we would ever get a model launch out of them, as the broader open model ecosystem did not slow down one bigt for them. Well, we did:
They compare themselves to Inkling, Nemotron, and GLM 5.2, but the SOTA GLM 5.3, Kimi K3, Qwen 3.8 Max, and DeepSeek V4.1 Flash are generally ahead. Still, since this is US-trained from scratch, there’s a segment of the market that has been eagerly waiting for more options here, and more importantly, Reflection has now announced its arrival as a functional neolab!