Mercor Raises $750 Million in Series D at $20 Billion Valuation

By Amara Osei

AI data company Mercor said it raised $750 million in Series D funding at a $20 billion valuation. Mercor said the financing will accelerate its expansion into enterprise AI evaluation and post-training infrastructure. That expansion targets businesses developing AI agents that must perform reliably within specific operating environments.

Founded in 2023 and headquartered in San Francisco, Mercor develops infrastructure connecting AI developers with specialists who evaluate and improve model performance. The company has traditionally worked with companies developing leading AI models, providing experts for training and evaluation work intended to improve system accuracy and usefulness.

Mercor said its network includes more than five million experts. Its customers include frontier AI laboratories, major technology companies, AI startups and Fortune 500 businesses.

Mercor is now expanding its offerings to enterprises applying existing AI models to internal business processes. The company believes enterprise adoption will require systems tailored to each company’s procedures, policies and technical environments.

Why do enterprise agents need custom evaluation?

Brendan Foody, Mercor’s co-founder and co-CEO, said: “Every job in the economy will need its own AI agents, and every company will need a way to train and evaluate them on its own work. We spent the last two years building that infrastructure with the frontier labs. Now we’re bringing it to enterprises across the economy.”

Mercor says general-purpose models can score well on public benchmarks yet struggle inside corporate systems. Business workflows often contain specialized terminology, proprietary information, compliance requirements and exceptions absent from standard testing environments.

Adarsh Hiremath, Mercor’s co-founder and co-CEO, said: “Every company deploying AI needs to know whether it works, and the answer is different for each one. A model can score well on a public benchmark and still fail on the systems, policies, and edge cases that define how a specific business runs.”

“We build custom benchmarks with each enterprise to measure models against its own standards, and that benchmark becomes the starting point for post-training,” Hiremath said. Mercor develops customized evaluations that measure how reliably agents complete particular tasks and what those tasks cost.

How does Mercor test agents before deployment?

Mercor’s evaluation platform establishes measurable performance standards before organizations deploy agents more broadly. The company believes those standards help determine whether AI systems are ready to manage business-critical activities.

The company also supports post-training, which adapts an existing AI model to improve performance against selected requirements. Enterprises can refine AI systems using information about their own operations. Mercor is building technology to evaluate and adapt existing models, rather than focusing exclusively on developing new foundation models.

Earlier in 2026, Mercor acquired Deeptune, which developed simulated training environments based on enterprise software applications. The acquired team spent approximately two years recreating hundreds of business applications as environments for AI training.

These simulations let AI agents practice completing workflows in settings resembling actual enterprise systems. Mercor plans to combine that capability with its expert network and evaluation technology.

Mercor’s approach builds on APEX, a public benchmark series designed to measure AI performance on economically valuable professional tasks. The company expects increasing demand from businesses seeking to deploy AI agents in specialized roles.

Who is backing the round, and what will the money fund?

Mercor named the investors in the $750 million Series D:

• Benchmark, Felicis, Fundomo, funds managed by affiliates of Apollo, G Squared, General Catalyst and Glade Brook Capital.
• NVIDIA, Salesforce, Sapphire Ventures, StepStone Group, XTX Ventures and Zoom.

The financing will support additional research, development of evaluation methodologies and expansion of post-training infrastructure. Mercor said the $750 million provides substantial resources for expanding the business.

Jai Das, Sapphire Ventures’ president, partner and co-founder, said: “For the last few years, frontier labs have been the primary buyers of expert-driven training data, which is about to change. As open models become good enough to build on, every enterprise will want to evaluate and train agents on its own work.”

Das said: “Mercor has built something hard to replicate—an expert network, task design, as well as training and evaluation environments shaped by several years of working with the frontier labs.” “We’re thrilled to partner with Brendan, Adarsh, and team to help them scale in the Enterprise market,” Das said.

The investment reflects investor interest in infrastructure supporting broader enterprise AI adoption, Mercor said. Finance, healthcare, law, engineering and other professional services often require systems capable of navigating complex, domain-specific tasks, according to the company.

Mercor believes combining human expertise with realistic training environments can help AI models improve in those sectors. At $20 billion, the company is positioning itself as foundational infrastructure for enterprises deploying AI systems trained on their own workflows and professional expertise.

The commercial test for that valuation will be whether enterprises adopt Mercor’s custom benchmarks and post-training systems as agents take on specialized roles.