
Nvidia has acquired enterprise AI startup Kumo AI for at least $400 million, strengthening its push into predictive foundation models for structured business data, reported The Information. The deal brings Kumo’s co-founders and proprietary graph neural network technology into Nvidia’s enterprise AI portfolio.
Kumo AI is a Mountain View–based enterprise AI startup founded in 2021 by Vanja Josifovski, Hema Raghavan, and Jure Leskovec. It raised $37 million in funding, led by Sequoia Capital, before being acquired by Nvidia in 2026.
In terms of funding, Kumo AI raised $37 million in 2022, with Sequoia Capital leading the investment. Other backers included SV Angel and a group of angel investors. Before its acquisition by Nvidia in 2026, Kumo had already secured enterprise customers such as Reddit, Sainsbury’s, DoorDash, Databricks, and Snowflake, positioning itself as a key player in predictive AI for business applications.
Key Facts About the Acquisition
- Deal Value: At least $400 million (reported by The Information).
- Company Acquired: Kumo AI, founded in 2022, focused on predictive AI for enterprise data.
- Co-founders: Vanja Josifovski, Hema Raghavan, and Jure Leskovec (all now at Nvidia).
- Funding History: Raised $37 million in 2022 from Sequoia Capital.
- Customers: Reddit, Sainsbury’s, DoorDash, Databricks, Snowflake.
- Technology: Uses graph neural networks and a proprietary Predictive Query Language to automate data preparation and answer predictive business questions.
Strategic Purpose
- Enterprise AI Expansion: Nvidia is broadening beyond chips into enterprise AI software, embedding predictive models directly into workflows.
- Hardware Synergy: Kumo’s models are expected to be optimized for Nvidia GPUs, strengthening Nvidia’s AI foundry software ecosystem.
- Competitive Edge: Helps Nvidia compete with hyperscalers (AWS, Google Cloud, Microsoft) and enterprise-focused AI rivals like Anthropic.
- Use Cases: Customer churn prediction, payment risk analysis, and operational forecasting for structured enterprise data.
Technical Highlights
- Graph Neural Networks: Represent enterprise data as nodes and edges, enabling richer predictive insights than row-based models.
- Predictive Query Language: Translates natural-language questions into model queries.
- Automation: Cuts manual data preparation effort by up to 95%.
- Latest Release: KumoRFM-2 model (April 2026), designed for structured data analysis.
Nvidia’s Broader Context
- Q1 FY2027 Revenue: $81.6B (up 85% YoY), driven by data center demand.
- New Products: Vera CPU, Cosmos 3 foundation model for robotics/autonomous driving.
- Challenges: Component shortages limiting near-term growth despite record demand.
Why It Matters for Enterprises
- Faster AI Deployment: Kumo’s automation reduces setup time, making predictive AI accessible to non-technical teams.
- Integration Potential: Likely to be embedded into Nvidia’s enterprise AI stack, enabling customized models for industries like retail, finance, and logistics.
- India Context: Gurugram’s IT and consulting firms could benefit early, given Nvidia’s global rollout and Kumo’s enterprise focus.
Nvidia’s Major Recent Acquisitions (2024–2026)
| Company | Year | Deal Value | Sector | Key Focus |
|---|---|---|---|---|
| Run:ai | 2024 | $700M | AI Infrastructure | Israeli AI orchestration platform for GPU scheduling and enterprise AI workloads. |
| Deci | 2024 | $300M | Deep Learning | Acceleration platform improving model performance up to 15x. |
| Groq | 2025 | $20B | AI Chips | High-performance inference processors; Groq’s TPU-like tech integrated into Nvidia’s AI factory. |
| Shoreline.io | 2024 | $100M | Cloud Ops | Automates system incident fixes; founded by ex-AWS executive. |
| OctoAI | 2024 | $250M | Healthcare AI | Seattle-based startup focused on medical AI solutions. |
| VinBrain | 2024 | Undisclosed | Healthcare AI | Vietnamese healthcare AI subsidiary of Vingroup. |
| Kumo AI | 2026 | $400M | Predictive AI | Structured data predictive models using graph neural networks. |
| OpenAI (equity stake) | 2026 | $30B+ | Generative AI | Strategic investment to deepen Nvidia’s role in LLM deployment. |
| Corning | 2026 | $2.7B | Optical Connectivity | Strengthening data center interconnects. |
| Intel (stake) | 2025 | $5B | Semiconductors | Partnership to expand AI chip collaboration. |
IndianWeb2.com is an independent digital media platform for business, entrepreneurship, science, technology, startups, gadgets and climate change news & reviews.
No comments
Post a Comment