Reflection launches Beam, an open AI model to compete with Chinese models at lower computational cost

Reflection AI officially unveils Beam, its first open, frontier AI model. The two-year-old Brooklyn-based startup claims that Beam matches the performance of leading Chinese open models on advanced reasoning benchmarks at significantly lower costs, a claim that could accelerate the race to build a Western answer to DeepSeek, Qwen and Z.ai.

Reflection’s announcement confirms reports from Axios over the weekend that the startup was close to launching. The company shared new details in a lengthy blog post Monday, which described Beam as a text-only expert mixture model trained in high-computation reinforcement learning to be effective in reasoning, coding, and agent tasks at “a fraction of the token cost and computational time” of competitors.

Beam is a 501 billion parameter model with 23 billion active parameters. It has been pre-trained on 23.8 trillion tokens and has a popup of 1 million tokens. For comparison, Z.ai’s GLM-5.2 has around 744 billion parameters in total, of which 40 billion are active.

Reflection’s claimed performance has not been independently verified, but based on advanced reasoning testing, the company claims that Beam achieves results comparable to Z.ai’s GLM-5.2 and outperforms current leading Western open models while using “3-4x less inference compute.” Reflection calls it a “workhorse model” for businesses, the public sector and developers.

Reflection positions itself against closed labs like Anthropic and OpenAI, against popular open models from Chinese developers, and against Western players like Mistral, Meta, and Cohere. Its most direct American rival might be Inkling, the open model from Mira Murati’s Thinking Machines Lab released in July. Reflection’s own tests show that Beam outperforms Inkling on four coding tests where both report results, but Inkling is a multimodal model and Beam is text-only.

Reflection was founded in 2024 by two former Google DeepMind researchers and has raised about $4.7 billion from backers including Nvidia, Sequoia Capital and Lightspeed Venture Partners, according to PitchBook. Its latest funding round valued the company at a pre-money valuation of $25 billion.

The startup also locked down compute – a key ingredient needed to train frontier models that can lure customers away from the closed models of Anthropic and OpenAI, as well as cheaper open models from Chinese labs. This summer, Reflection signed deals collectively worth more than $7 billion with SpaceX and Nebius to secure access to Nvidia’s GB300 chips through 2029.

The reflection targets Beam and future models towards businesses and sovereign nations. The goal is to create “AI factories,” a product that would allow institutions to create their own local, custom AI system by training Reflection’s AI models on their own proprietary data. Nvidia CEO Jensen Huang, whose company backs Reflection, has long championed the idea of ​​the “AI factory” and pushed for strengthening the open AI ecosystem — a vision that would also benefit Nvidia, whose GPUs would power those systems.

Axios reported that hedge funds and trading firms are among those interested in creating such systems. Reflection has already begun testing the concept of a sovereign AI factory partnership with South Korea’s Shinsegae Group.

Reflection says it will release weights and full technical details for Beam this month, with distribution via hyperscalers and neoclouds and integrations into open source libraries at launch.

Reflection did not respond in time to TechCrunch’s requests for more information.