Mercor raises $750 million to expand AI training and workplace evaluations
The $20 billion valuation comes as Mercor takes its human-expertise business deeper into enterprise agents.
Mercor announced a $750 million Series D on November 10 at a $20 billion valuation, as the AI training company expands its work helping businesses test and adapt agents for professional tasks. Reuters reported the financing the same day. Mercor · Reuters
The round included an employee tender, according to Mercor. Investors included Benchmark, Nvidia, General Catalyst, Salesforce and funds managed by Apollo affiliates. The announcement does not break out how much of the headline amount represents new capital for the business versus employee liquidity. Mercor
Mercor began as an AI-assisted hiring platform and now connects companies with specialists whose knowledge helps train and assess models. Its valuation was about $10 billion after a $350 million financing in October 2025, according to Reuters. Reuters
The new expansion centers on evaluations built from a company’s actual workflows, further training of models and simulated environments where agents can practice. Mercor says it wants enterprises to adapt capable open models using their own domain knowledge and operating requirements. These are its intended uses of the funding, rather than independently demonstrated customer results. Mercor
The business opportunity is in the work between buying a model and trusting it with a job. An agent can perform well on a public benchmark yet struggle with an employer’s processes, permissions or quality standards. Company-specific tests can expose those gaps before a wider deployment.
For buyers, useful evidence will be performance on representative tasks, error rates, costs and the amount of human checking still required. The funding gives Mercor more resources to build that evaluation business; the valuation alone says little about how reliably any particular agent will work.
Takeaway: The investment targets company-specific agent evaluation and training; the employee-tender component and customer outcomes remain important distinctions.
Sources and citations
Revision history
Published .