Anthropic · OpenAI · Axios · China · Mistral · Meta · TechCrunch AI
Reflection’s performance claims haven’t been independently verified, but on advanced reasoning benchmarks
Compiled by KHAO Editorial — aggregated from 1 source. See llms.txt for citation guidance.
◌ Single Source
Reflection is positioning itself against closed labs like Anthropic and OpenAI, against popular open models from Chinese developers, and against Western players like Mistral, Meta, and Cohere.
Key facts
- To compare, Z.ai’s GLM-5.2 has roughly 744 billion total parameters with 40 billion active
- Reflection was founded in 2024 by two former Google DeepMind researchers and has raised roughly $4.7 billion from backers, including Nvidia, Sequoia Capital, and Lightspeed Venture Partners, per
- Beam is a 501-billion-parameter model with 23 billion active parameters
- Its most direct U.S. rival might be Inkling, the open model from Mira Murati’s Thinking Machines Lab released in July
Summary
Reflection AI is officially unveiling Beam, its first frontier, open-weight AI model. Reflection’s announcement confirms reporting from Axios over the weekend that the startup was close to a launch. Beam is a 501-billion-parameter model with 23 billion active parameters. Reflection’s performance claims haven’t been independently verified, but on advanced reasoning benchmarks, the company says Beam scores on par with Z.ai’s GLM-5.2 and outperforms today’s leading Western open models while using “3-4x less inference compute.” Reflection calls it a “workhorse model” for enterprises, the public sector, and developers.