The news
Harvey, a legal AI company, said on September 9, 2026, that it raised $550 million at a $15.5 billion valuation, co-led by Diffusion and Lightspeed Venture Partners, to help law firms and legal departments build their own AI.
Sequoia, Kleiner Perkins, Andreessen Horowitz, Coatue, Conviction, Elad Gil, Evantic, GIC, Goldman Sachs Alternatives, Whale Rock, WndrCo, Sapphire Ventures and Verified Capital also invested, according to Harvey. TechCrunch reported that the new valuation compares with $11 billion in March 2026 and $8 billion in December 2025, and both TechCrunch and LawSites put Harvey's total funding to date at more than $1.5 billion.
Harvey said it works with 80% of the Am Law 100, a ranking of the largest US law firms, and with in-house legal teams at five Fortune 10 companies. Legal technology outlet LawSites reported that the company has more than 3,000 customers and annual recurring revenue (ARR) above $400 million.
Two product moves framed the raise. Harvey introduced Tenet, which TechCrunch described as an in-house model built by post-training the open-weight Kimi K3 model on legal data, using inference provider Fireworks. Post-training means adapting an existing model with specialist data, and open-weight models are those whose parameters customers can download and run. Harvey also launched Harvey LAB, a Legal Agent Benchmark for measuring how AI agents perform on legal work.
Co-founder and chief executive Winston Weinberg wrote on LinkedIn, as reported by LawSites, that the money will go to “people and compute,” meaning aggressive hiring across product, engineering, research and go-to-market roles, plus investment in model training and compute. The company framed its mission as helping legal teams build and own their intelligence at scale.
The numbers
- Round size
- $550 million
- Valuation
- $15.5 billion
- Valuation in March 2026
- $11 billion (TechCrunch)
- Total raised
- More than $1.5 billion (TechCrunch, LawSites)
- Annual recurring revenue
- Above $400 million (LawSites)
- Customers
- More than 3,000 (LawSites)
Why CEOs should care
For general counsel and law firm leaders, Tenet signals a change in how legal AI is sold. Instead of renting a general-purpose model through a vendor, Harvey is pitching customers on adapting models with their own data and owning the result. Before signing, ask who owns the fine-tuned model, whether it can be exported if the contract ends, how client-confidential material is isolated during training, and which benchmark results the vendor will stand behind in writing.
For CFOs, ARR above $400 million across more than 3,000 customers indicates legal AI has moved beyond experiments into budgeted spending. That scale gives buyers leverage. Renewals should be tied to usage reporting and outcome measures, such as hours saved per matter or turnaround time on contract review, rather than seat counts that go unused.
For boards of law firms and professional services companies, the valuation jump is a competitive signal. If rivals embed AI agents into document review and drafting, pricing built on billable hours comes under pressure. Directors should ask management how AI productivity gains will be shared with clients, and how the firm will stand apart if competitors license the same tools.
The bigger picture
Harvey's valuation has nearly doubled in about nine months, based on TechCrunch's figures, and it has closed five priced rounds since 2025. Its closest rival is Legora, a Swedish company that raised $550 million at a $5.55 billion valuation in March, led by Accel, according to Crunchbase News. Both are also acquirers: Crunchbase counted at least five acquisitions by Legora and at least three by Harvey in 2026 as of late August, with prices undisclosed.
The wider legal technology market is running below its 2025 peak. Crunchbase tallied $2.2 billion raised by legal tech startups in 2026 through late August, compared with a record $4.6 billion for all of 2025. The large rounds are going to a handful of leaders, a pattern Crunchbase has documented across venture capital this year as dollars concentrate in fewer companies.
What’s next
Watch whether large law firms adopt Tenet-style custom models or stay with general-purpose tools, and whether Harvey LAB becomes a benchmark buyers cite in procurement. Also watch Harvey's acquisition pace and hiring: Weinberg said the new capital is aimed at people and compute, so product breadth and model quality are the measures to track before the next renewal cycle. Buyers comparing Harvey with Legora should ask both for customer references at firms of similar size and practice mix.
What “Fact-checked” means
Fact-checking means testing a story’s facts against the evidence before it is published. This story went through at least two separate checks before this version was published.
- What we checked
- Its names, figures, dates, job titles, quotes and who said what were checked against the story’s sources, including its main source where it could be opened. The headline was checked for accuracy and overstatement.
- How
- A first check reviewed the whole story. If it passed, a second, skeptical check went back to the sources to look for mistakes in the most important facts. If a check flagged the story, it was edited to fix the problems found, and a separate re-check then reviewed the whole story again.
- Who
- The checks are made with our newsroom’s technology tools, as steps kept separate from the writing, under rules set by our editor, Hussein Mukhtar. A story the checks still flag is not published automatically; it is held for the editor, who decides whether it is fixed, published or dropped.
- If something is wrong
- “Fact-checked” does not mean error-free. If a material error is found after publication, we correct the story and add a note saying what changed. Report an error
