Legal AI start-up Harvey raises $550m at $15.5bn valuation

Six months after its last multi-million dollar funding round, AI startup Harvey AI Corp. has raised an additional $550 million at a valuation of $15.5 billion. The investment was led by diffusion and Lightspeed Venture Partners, with additional backing from over a dozen other venture capital firms, including Sequoia, Kleiner Perkins, and Goldman Sachs.

Harvey specializes in providing law firms with a cloud-based platform to automate manual tasks for attorneys. The company's installed base comprises 80% of the 100 highest-ranked law firms in the United States, and major enterprises, such as half the Fortune 10, also utilize their services.

The platform enables users to store up to 100,000 documents in a repository called Vault. An integrated search engine utilizes AI to identify useful patterns, allowing attorneys to efficiently locate and modify supplier contracts that require compliance with new regulations.

Harvey's AI agents assist in automating complex tasks by retrieving external data like precedents and legislative clauses. These agents prompt attorneys for input when clarification is needed for intricate parts of a task. The company recently debuted its first custom large language model, Tenet, which is based on the open-source Kimi K3 model with 2.8 trillion parameters.

Fine-tuned to suit legal-related tasks, Tenet comprises 896 individual neural networks, each optimized for distinct functions. Harvey claims that Tenet outperforms both Kimi K3 and two other popular models, Fable 5 and GPT-5 Sol, in various aspects, achieving improved contract processing speeds by up to 20%. The company aims to further enhance its AI capabilities by investing in the development of more generalist models.