AI Training Startup Mercor Raises $750 Million at $20 Billion Valuation

Mercor, a San Francisco-based artificial intelligence training startup, has raised $750 million in a Series D funding round at a valuation of $20 billion. The latest investment highlights growing demand for companies that help develop, train and evaluate advanced AI models using human expertise. The funding round attracted major technology companies and venture capital investors, including NVIDIA, Salesforce, Zoom, Benchmark and General Catalyst.

Founded in 2023, Mercor initially operated as an AI-powered hiring platform before expanding into the business of supplying specialised human expertise for AI training. The company connects AI developers with professionals who help train and evaluate models across fields such as finance, law, medicine, science and software engineering. Mercor plans to use the new capital to expand its AI evaluation and post-training infrastructure, helping businesses build AI agents capable of performing complex professional tasks.

Mercor Raises $750 Million in Series D Funding

Mercor announced the $750 million Series D financing on November 10, 2026, putting its valuation at $20 billion.

The round attracted a combination of existing venture capital backers and major technology companies. Investors participating in the financing include:

• Benchmark
• Felicis
• Fundomo
• Apollo-affiliated funds
• G Squared
• General Catalyst
• Glade Brook Capital
• NVIDIA
• Salesforce
• Sapphire Ventures
• StepStone Group
• XTX Ventures
• Zoom

The participation of NVIDIA, Salesforce and Zoom is particularly notable because these companies operate in different parts of the technology ecosystem. Their investment reflects growing interest in the infrastructure required to make AI systems useful in real-world business environments.

The latest valuation represents a substantial increase from Mercor’s previous $10 billion valuation, which followed a $350 million Series C funding round in October 2025.

The new financing also places Mercor among a growing group of highly valued private AI companies attracting significant institutional and corporate investment.

What Does Mercor Do?

Mercor operates in the AI data and training industry, helping technology companies improve the performance of artificial intelligence models.

Modern AI systems require large amounts of high-quality data and human feedback to become more capable. While publicly available information can help models learn general knowledge, specialised professional tasks often require more detailed examples and expert evaluation.

Mercor addresses this requirement by connecting AI developers with professionals who possess expertise in specific industries.

For example, a financial expert may evaluate whether an AI model can correctly interpret financial documents, while a lawyer may assess its ability to understand legal language. Medical professionals, scientists and software engineers can similarly contribute specialised knowledge.

This process helps AI developers identify weaknesses in their models and improve their ability to complete complex tasks.

Mercor says its network includes more than five million experts who help train and evaluate AI models. Its customers include frontier AI laboratories, major technology companies, AI startups and large enterprises.

Mercor Expands Beyond Data Labelling Into AI Agent Training

Mercor’s business is evolving beyond traditional data labelling and annotation services.

As AI models become more capable, companies increasingly want AI agents that can perform complete workflows rather than simply generate text or answer questions.

An AI agent may need to analyse a spreadsheet, update a customer relationship management system, prepare a report or complete a series of tasks across multiple business applications.

Training such systems requires more than providing examples of correct answers. Developers need realistic environments in which AI agents can practise tasks, receive feedback and improve their performance.

Mercor is investing in infrastructure that combines three important capabilities:

1. Human expertise: Specialists create tasks, assess results and provide feedback based on professional knowledge.
2. AI evaluation: Custom benchmarks measure whether models can complete specific tasks accurately and reliably.
3. Post-training infrastructure: Additional training helps improve an existing model’s ability to perform specialised work.

The company’s objective is to help businesses determine whether an AI agent can perform a particular job, how much it costs to operate and where further training is required.

Deeptune Acquisition Strengthens Mercor’s AI Infrastructure

Mercor’s expansion into AI agent training received an additional boost through its acquisition of Deeptune earlier in 2026.

Deeptune specialises in creating simulated environments that allow AI agents to practise professional workflows.

The company has recreated hundreds of enterprise applications, including software used for spreadsheets and customer relationship management. These environments allow AI agents to interact with software and attempt tasks similar to those performed by human employees.

The training process involves three elements:

• Software environments: Digital versions of applications in which AI agents perform tasks.
• Tasks: Clearly defined activities that an AI agent must complete.
• Verifiers: Systems that determine whether the agent successfully completed the assigned task.

For example, an AI agent could be asked to update information in a simulated business application, prepare a spreadsheet or complete a customer-service workflow. The system then checks whether the result meets the required standards.

The acquisition strengthens Mercor’s ability to combine human expertise with realistic software environments, an important requirement for training more capable AI agents.

Why NVIDIA, Salesforce and Zoom Are Investing in Mercor

The investor list highlights the growing strategic importance of AI training infrastructure.

NVIDIA is a major supplier of GPUs and computing infrastructure used to develop and operate advanced AI models. High-quality training data and evaluation systems are important components of the broader AI development ecosystem.

Salesforce operates enterprise software platforms used for customer relationship management, sales, marketing and business operations. AI agents capable of working reliably within enterprise applications could create new opportunities for the company’s software ecosystem.

Zoom, meanwhile, has been expanding its AI capabilities across communication and workplace productivity products. Improvements in AI training and evaluation could support the development of more capable workplace assistants.

The investments do not necessarily indicate that every participating company has entered into a commercial partnership with Mercor. However, their participation demonstrates the strategic interest surrounding companies that provide the data, expertise and infrastructure needed to develop useful AI systems.

Mercor’s Growth Reflects the Expanding AI Training Market

Mercor’s valuation increase comes as investment in artificial intelligence startups continues to accelerate.

According to Crunchbase figures cited by The Economic Times, AI-driven startups raised approximately $102 billion during the third quarter of 2026, representing 64% of global venture capital funding during the period.

The figures illustrate how investors are directing a growing share of venture capital towards companies developing AI models, infrastructure and applications.

The AI industry is also moving beyond the initial focus on building increasingly powerful models. Businesses now want to deploy these systems in areas where they can deliver measurable productivity improvements.

This shift creates opportunities for companies providing specialised training data, model evaluations and enterprise-specific AI environments.

Mercor is positioning itself within this market by helping companies move from general-purpose AI models towards systems capable of completing industry-specific work.

What the $20 Billion Valuation Means for Mercor

The latest funding round values Mercor at $20 billion, twice its previous valuation of $10 billion.

The increase reflects investor expectations about the future demand for AI training and evaluation services.

However, a private-company valuation is not the same as a publicly traded market capitalisation. It represents the valuation assigned to the company in a private financing transaction and may change depending on future funding conditions, business performance and investor demand.

The latest round also highlights the importance of infrastructure companies in the AI economy. While public attention often focuses on model developers and chatbot products, companies providing training data, expert feedback and evaluation systems are becoming increasingly important.

For Mercor, the challenge will be turning its rapid expansion into a sustainable business while maintaining the quality and reliability of its services.

The Bigger Picture

Mercor’s $20 billion valuation reflects a broader shift in the artificial intelligence industry. As AI models become more advanced, the focus is moving from simply building powerful systems towards ensuring that they can perform real professional tasks reliably. Human expertise, specialised training data and realistic evaluation environments are becoming important parts of that process.

The participation of NVIDIA, Salesforce, Zoom and major venture investors also demonstrates the growing strategic value of AI training infrastructure. Mercor is attempting to establish itself as a key provider in this emerging market by combining expert networks, model evaluation and agent-training environments. Its long-term success will depend on whether businesses increasingly adopt AI agents and pay for the infrastructure required to make them dependable.

Looking Ahead

Mercor plans to use its latest funding to accelerate research and expand its evaluation and post-training infrastructure. The company is also working to increase the number of realistic environments in which AI agents can learn complex workflows. Its expansion into enterprise customers could provide a larger market beyond the frontier AI laboratories that initially drove demand for specialised training data.

The next phase of Mercor’s growth will depend on how effectively it converts its expert network and technical infrastructure into recurring enterprise business. Competition from other AI data and training companies will remain significant, while customers will increasingly demand measurable improvements in AI performance. The $20 billion valuation sets high expectations, making execution and sustainable revenue growth important milestones for the startup.