Releases

An independent tracker counted 48 live postings on Reflection's hiring system. The count is a point-in-time external snapshot.

Reflection made the weights for its first model, Beam, publicly available, establishing its open model family.

Reflection released Beam, a 501B-total-parameter sparse mixture-of-experts model with 23B active parameters for coding, reasoning and agentic work.

Reflection described a planned AI Factory offering that would allow institutions to build localized AI ecosystems around its open-weight models.

Nebius agreed to supply approximately $1 billion of AI compute capacity to Reflection.

Reflection agreed to obtain NVIDIA GB300 compute capacity from SpaceXAI's Colossus 2 infrastructure under a contract reported to be worth up to $6.3 billion through 2029.

Reflection became an AI model provider for the Department of Energy's Genesis Mission, supporting customizable models for U.S. national laboratories.

Reflection's CEO confirmed the company had closed a $2.5 billion Series C financing to fund open-weight frontier-model development and compute commitments.

A third-party workforce dataset estimated 177 Reflection employees in the first quarter of 2026. This is not a company-reported count.

Reflection raised $2 billion in a financing led by NVIDIA with participation from Lightspeed, Sequoia and other investors.

Reflection announced a strategy centered on releasing frontier open models and building a scalable commercial model around customizable open intelligence.

Reflection disclosed a $105 million Series A led by Lightspeed Venture Partners and CRV, bringing the announced financing total to $130 million.

Reflection disclosed a $25 million seed financing led by Sequoia Capital and CRV.

Reflection publicly described its plan to pursue superintelligence through autonomous coding systems and reinforcement learning.