Sapiom, a San Francisco-based startup, has raised a $35 million Series A round to reduce the cost of running AI agents, sometimes by up to tenfold. The round was led by Dragonfly, with Anthropic among the backers.
TL;DR
- Sapiom raises $35M Series A to make AI agents cheaper to run, with Anthropic as an investor.
- The startup's Router directs calls to the cheapest capable model, reducing costs significantly.
- Sapiom is capitalizing on the shift towards cost-effective AI solutions as companies audit their AI budgets.
What happened
Sapiom, a startup that sits between AI agents and the models they run on, has raised $35 million in Series A funding. The round was led by Dragonfly, with participation from Anthropic, Okta Ventures, Menlo Ventures, and Array Ventures. This brings Sapiom's total funding to $50 million, following a $15 million seed round six months ago.
Sapiom's platform processes requests from AI agents and directs them to the most cost-effective model or service. Its Router ensures that each call is sent to the cheapest capable model, not necessarily the most expensive one. Since its launch six months ago, the platform has processed over 270 million transactions.
One of Sapiom's customers, Polsia, saw its monthly token bill drop from $1.2 million to about $100,000 after implementing Sapiom's solution. This significant reduction underscores the startup's value proposition.
Why it matters
Sapiom is addressing a critical pain point for AI startups: the high cost of running AI agents. As companies increasingly audit their AI budgets, solutions that reduce costs are in high demand. Gartner forecasts that over 40% of agentic AI projects may be canceled by the end of 2027 due to escalating costs.
The involvement of Anthropic, a leading AI model maker, as an investor highlights the strategic importance of Sapiom's solution. While it may seem counterintuitive for a model maker to invest in a startup that reduces spending on models, Sapiom's founder argues that cheaper inference will lead to more widespread adoption of AI agents, benefiting model makers in the long run.
Sapiom faces competition from established players like OpenRouter and cloud providers like Amazon and Microsoft, which are bundling routing services into their offerings. However, Sapiom's infrastructure-based approach and direct compute charging may give it an edge in the crowded market.
Key facts
- Sapiom raises $35 million Series A, led by Dragonfly.
- Total funding reaches $50 million, including a $15 million seed round.
- Platform processes over 270 million transactions since launch six months ago.
- Customer Polsia reduces monthly token bill from $1.2 million to $100,000.
- Anthropic, Okta Ventures, Menlo Ventures, and Array Ventures participate in the round.
- Sapiom's Router directs calls to the cheapest capable model.
- Gartner predicts over 40% of agentic AI projects may be canceled by the end of 2027 due to costs.
- OpenRouter moves around 25 trillion tokens a week.
Context
The AI industry is undergoing a shift towards cost-effective solutions as companies grapple with the high expenses of running AI models. Startups like Sapiom are emerging to address this need by optimizing the use of AI resources.
The involvement of major investors like Anthropic and venture capital firms highlights the strategic importance of cost reduction in the AI landscape. As more companies adopt AI agents, the demand for solutions that make these agents more affordable will continue to grow.
Sapiom's infrastructure-based approach sets it apart from competitors that act purely as middlemen. By owning the inference and controls around it, Sapiom aims to differentiate itself in a crowded market.
