Reflection AI, a startup founded by former Google DeepMind researchers, is preparing to release an open-weight foundation model to compete with Chinese models DeepSeek and Alibaba’s Qwen, which currently dominate many open-source model leaderboards. According to an Axios report dated October 4, 2026, the model is expected to arrive this month. Founded in March 2024 by Misha Laskin and Ioannis Antonoglou, the startup raised $2 billion at an $8 billion valuation in October 2025, with current valuations reportedly ranging between $8 billion and $25 billion. It has struck a compute deal with Nebius worth more than $1 billion and secured substantial capacity from SpaceX’s Colossus, including significant Nvidia GPU rentals. The company is betting on an AI factory model that combines open weights with client proprietary data and advanced Nvidia GPU compute to deliver customized, localized AI deployments at lower cost. For enterprises, a model that genuinely matches the top Chinese open-weight systems would provide another credible option that they can host, customize and control.
Source: Read the original article
