Editorial illustration for Reflection Signs USD 1 Billion Compute Deal With Nebius
Reflection AI Lands $1B Compute Deal With Nebius
Reflection Signs USD 1 Billion Compute Deal With Nebius
Reflection AI has locked in a $1 billion compute deal with Nebius, giving the open-model startup access to Nvidia's newest chips through the European infrastructure company. The agreement lands just weeks after Reflection signed a separate arrangement to tap SpaceX's computing capacity, part of a broader scramble among AI developers to secure the hardware needed to train and run their models.
Founded in 2024 by two former Google DeepMind researchers, Reflection is now valued at $8 billion and has raised roughly $2.6 billion from investors including Nvidia, Sequoia Capital, and Lightspeed Venture Partners. The company has positioned itself among a growing group of open-weight model builders drawing attention as scrutiny of closed-source AI intensifies, particularly after the Trump administration pushed Anthropic and OpenAI last month to restrict some of their most capable models.
For Nebius, formerly Yandex's international arm, the Reflection deal adds to a run of major contracts. The company recently signed a five-year, up to $27 billion infrastructure deal with Meta and a multi-year, up to $19.4 billion agreement with Microsoft last year.
Nebius, formerly the international arm of Russian tech giant Yandex, will provide Reflection access to Nvidia’s latest chips. The deal comes just a few weeks after the startup signed a similar deal to access SpaceX’s computing resources, and mirrors several partnerships by AI firms as they race to secure compute for training and deploying their models.
Why this matters For developers and founders betting on open-weight models, Reflection's back-to-back deals with SpaceX and now Nebius signal that compute access, not model architecture, is becoming the real competitive moat. A $1 billion commitment from a former Yandex subsidiary to a U.S. open-model startup also shows how global the infrastructure race has become; the money and hardware don't care much about borders once Nvidia chips are involved.
For researchers watching the closed-versus-open debate, this deal adds weight to the open-weight camp at a moment when data retention and sovereignty concerns are pushing enterprises to look past OpenAI and Anthropic. But we'd flag a real question: stacking multiple billion-dollar compute deals in a matter of weeks raises the stakes for Reflection to actually ship something competitive with its Chinese open-weight rivals, who've moved fast and cheap. Access to Nvidia's latest chips is necessary, not sufficient.
Watch whether Reflection turns this hardware into a model that developers actually adopt, rather than another well-funded promise.
Related Reading
Business & Startups Kling launches Video O1, all-in-one model with MVL bridge using transformer
Business & Startups Enterprise voice AI splits into three architectures, shaping compliance
Business & Startups Nordic pilot adds Gemini for Education, NotebookLM to boost AI literacy
From Archives 1Password expands into AI cost management as model pricing diverges
From Archives PixVerse Hits USD 2B Valuation With 150M Registered Users
Common Questions Answered
What is the value of Reflection AI's compute deal with Nebius and what hardware does it provide access to?
Reflection AI has secured a $1 billion compute deal with Nebius that gives the startup access to Nvidia's newest chips through the European infrastructure company. This agreement represents a major commitment of computing resources for training and deploying Reflection's open-weight models.
Why did Reflection AI sign deals with both SpaceX and Nebius for computing capacity?
Reflection AI signed back-to-back deals with SpaceX and Nebius as part of a broader scramble among AI developers to secure the hardware needed to train and run their models. These dual agreements demonstrate that compute access has become the critical competitive advantage in the AI industry, more important than model architecture alone.
Who founded Reflection AI and when was the company established?
Reflection AI was founded in 2024 by two former Google DeepMind researchers. The startup has quickly reached an $8 billion valuation despite being newly established, reflecting investor confidence in its open-model approach and compute strategy.
What is Nebius and why is its partnership with Reflection AI significant for global AI infrastructure?
Nebius is the former international arm of Russian tech giant Yandex that now operates as an independent European infrastructure company. The partnership between Nebius and U.S.-based Reflection AI demonstrates how global the infrastructure race has become, showing that compute resources and Nvidia chips transcend national borders in the competitive AI landscape.
Further Reading
- SpaceX inks compute deal with Reflection AI, an open source AI lab - TechCrunch
- AI startup Reflection signs computing power deal with SpaceX - Reuters
- Nvidia-backed Reflection lands SpaceX compute deal - Axios
- SpaceX signs computing power deal with open-source AI startup Reflection worth up to $6.3 billion - CNBC
- SpaceX Fuels AI Ambitions With $6.3 Billion Reflection Compute Deal - Benzinga
