01What it sells and who buys it
Reflection develops open models and the software and infrastructure to customize and run them. It targets enterprises and public organizations that want control over their AI environments.
02How the business makes money
Its official offering combines models, operating infrastructure and tailored solutions. Public materials do not disclose standard prices or contract revenue, so those cannot be treated as established subscription figures.
03Marketing and customer acquisition
Reflection emphasized open models and runtime control to attract researchers and enterprise partners.
- 01
Propose AI under enterprise control
SourceThe company develops models, customization and compute infrastructure, positioning controlled deployment for enterprise and public-sector buyers.
- 02
Financing the next stage
SourceReflection announced $2 billion in financing in October 2025 at a reported $8 billion valuation. Backers included NVIDIA, Sequoia and Lightspeed, supporting its models and training infrastructure.
- 03
How distribution and adoption expanded
SourceResearch, developer support and an open-model strategy introduce the platform. Enterprise material emphasizes deployment and control on customer infrastructure. Research and partnership updates provide an observable prospect touchpoint, without a disclosed conversion rate.
How the growth mechanism works
- Publish the research direction
- Recruit developers and partners
- Evaluate controlled deployment
- Develop enterprise requirements
Research and update subscriptions are early relationship-building channels. They should not be described as mass paid adoption; investor research expectations and commercial deployment remain distinct.
04Funding and expansion
Reflection announced $2 billion in financing in October 2025 at a reported $8 billion valuation. Backers included NVIDIA, Sequoia and Lightspeed, supporting its models and training infrastructure.