Reflection Releases Open-Weight Model Beam, Focused on Low Inference Cost
Understand the story
In the early hours of October 6, 2026, US startup Reflection AI released its first frontier open-weight model, Beam, claiming it performs on par with the Chinese open-source model GLM-5.2 from Z.ai on advanced reasoning benchmarks while using 3-4 times less inference compute. Later that day, The Decoder reported more model details: Beam is a free open-weight model with 501B total parameters, uses a MoE architecture, activates 23B parameters per token, and is primarily aimed at coding, logical reasoning, and agent tasks. According to Reflection, Beam matches GLM 5.2 on difficult reasoning tasks with three to four times less compute, and approaches Qwen3.8-Max on coding and agent benchmarks. However, the report also notes that Beam's raw performance still falls short of stronger open-source models such as Kimi K3. The event is at an early release stage; independent evaluations of the model's actual performance have not yet been reported.
AI overview of reporting · Updated 11 days ago
Reporting timeline
Explore different aspects of the story through its reporting.
- The DecoderReflection's Beam becomes the most capable open-weight model built outside China
Reflection announced its first free open model, Beam, with 501B total parameters and 23B activated per token (MoE), aimed at coding, logical reasoning and agentic tasks. According to Reflection, Beam matches GLM 5.2 on hard reasoning tasks with three to four times less compute, and comes close to Qwen3.8-Max on coding and agentic benchmarks, though its raw performance still trails stronger open models like Kimi K3.
- TechCrunch · AIReflection debuts Beam, an open-weight AI model to rival Chinese models at lower compute cost
US startup Reflection AI officially released Beam, its first frontier open-weight model, claiming it performs comparably to Z.ai's GLM-5.2 on advanced reasoning benchmarks while using 3-4x less inference compute.