Active
I
Inferact
Promising · 37/100
Built by the creators of vLLM, making LLM inference faster and cheaper across any hardware
AI & MLSan Francisco, United StatesEst. 2026Seed · $150,000,000 · Valued at $800,000,000
0
Upvotes
7
Views
0
Feedback
30
Team Size
About
Inferact is founded and led by the core creators and maintainers of vLLM, the widely-used open-source inference engine that determines how efficiently AI models actually run once deployed. The company's thesis is that as models get more capable, the systems serving them — not the models themselves — become the real bottleneck, and it is building commercial infrastructure around vLLM to close that gap. vLLM itself now supports 500+ model architectures across 200+ accelerator types, with contributions from over 2,000 open-source community members, and in August 2026 became the first inference engine to complete full Kimi K3 vendor verification.
The founding team includes Simon Mo and Woosuk Kwon (vLLM's original creators), UC Berkeley professor Ion Stoica (also a co-founder of Databricks and Anyscale), Joseph Gonzalez, Roger Wang, and Kaichao You. Inferact raised $150M in a Seed round in January 2026 at an $800M valuation, led by Andreessen Horowitz and Lightspeed Venture Partners, with Sequoia, Altimeter, Redpoint, Databricks Ventures, and the UC Berkeley Chancellor's Fund among the other participants — an unusually large seed round reflecting how central vLLM already is to production AI infrastructure before Inferact even had a commercial product.
Business Score
37/100
Promising#2 in AI & ML
Score based on profile completeness, engagement & verification
Company Info
- Investors
- Andreessen HorowitzLightspeed Venture PartnersSequoia CapitalAltimeter CapitalRedpoint VenturesZhen FundThe House FundLaude Ventures
- Employees
- 30
Tags
High Tech > AI InfrastructureNative AI > AI Infrastructure
Contact
Similar Startups
View allR
Ricursive
Developer of self-improving artificial intelligence systems focused on chip design
AI & ML0
R
Resolve AI
AI agents that triage production incidents and run on-call so engineers stop firefighting
AI & ML0
R
Recursive
Developer of algorithms for recursive self-improving artificial intelligence research systems
AI & ML0
T
TENEX
Provider of cybersecurity suite solutions
AI & ML0