VC & Startups/Nov 16 edition/Reporting & analysis
FundingValuationsM&AVenture data

FundingRounds, leads & terms

Harvey and Legora raise $1.85 billion across four 2026 rounds as wider legal tech funding falls short of 2025's record

Legal tech raised a record sum in 2025, and 2026 has not kept pace. Money is concentrating, though: Harvey and Legora took $1.85 billion across four rounds, trackers disagree on sector totals, and both leaders are buying smaller rivals.

From the Business Insider · Startups article of November 16, 2026 and cited sources: Two companies account for much of 2026's legal tech funding. [1] [4] [7] [8]

SOURCES[1] Business Insider: Is legal tech funding losing steam? businessinsider.com[4] Tech.eu: Legora Series D tech.eu[7] LawNext: Harvey raises another $550M lawnext.com

AI-assisted, not human-edited. Methodology Corrections Who owns this site

KEY FACTS
What happened
Two companies account for much of 2026's legal tech funding. [1] [4] [7] [8] [9] [10]
Original source
Business Insider · Startups article, independent perspective
Brief by
AT Pulse research desk AI-assisted; publisher and contact on the desk page
When
November 16, 2026 article published
Evidence
13 sources across 9 sites

Reach: not measured yet

Illustration from Business Insider: Harvey and Legora raise $1.85 billion across four 2026 rounds as wider legal tech funding falls short of 2025's record
Image: Business Insider — Original article ↗
THE CORE IDEAS3 TAKEAWAYS
01

Two companies account for much of 2026's legal tech funding. Legora raised $550 million at a $5.55 billion valuation in March and another $550 million at $8.8 billion post-money in October. Harvey raised $200 million at $11 billion and then $550 million at $15.5 billion in September. Both companies report fast revenue growth: Harvey says its annual recurring revenue is above $400 million, and Legora says above $200 million. These revenue figures are self-reported. [1] [4] [7] [8] [9] [10]

02

Whether the sector is slowing depends on the data provider. Crunchbase puts 2025 at a record $4.6 billion and counts $2.2 billion for 2026 through late August. Legal Complex puts 2025 at $5.99 billion. It also counts $2.34 billion for the first quarter of 2026 alone, a total that includes Relativity's $720 million debt facility. Crunchbase's figure also predates about $1.1 billion that Harvey and Legora raised later in the year. [1] [2] [3] [6]

03

The market is consolidating. Crunchbase counts at least three acquisitions by Harvey and five by Legora in 2026, and Wolters Kluwer bought Brightflag for $500 million. Beyond the leaders, many companies funded between 2020 and 2023 have not raised again, and 2025 exit value fell 39%. Early-stage money is still flowing, though: in the first quarter, seed deals outnumbered growth deals for the first time since early 2024. [2] [3] [6]

WHY IT MATTERS

Round sizes, valuations and acquisition counts all show capital pooling around two vendors. The likely result, not yet proven, is harder later-stage fundraising for rivals while seed funding continues. Buyers should expect fewer, larger suppliers.

Research is done. The dossier follows.

Executive brief

Legal tech funding has not lost steam. It has become concentrated in two companies. Business Insider reports that the sector raised a record $4.6 billion in 2025 and that 2026 has not kept pace (Business Insider). Even so, Harvey and Legora together raised $1.85 billion across four 2026 rounds. Harvey is now valued at $15.5 billion and Legora at $8.8 billion. Through late August, Crunchbase counted $2.2 billion for the whole sector in 2026 (Crunchbase News). That count came before about $1.1 billion more from these two companies alone. Whether the year looks weak or strong depends on which data provider's methods are used.

What changed and event timeline

  1. A different tracker counts $5.99B for 2025

    Using Legal Complex data, Artificial Lawyer reported $5.99 billion raised in 2025, up 22%. The number of funded companies fell 27% to 292 ().

  2. Legora's Series D

    Legora raised $550 million at a $5.55 billion valuation in a round led by Accel. The valuation roughly tripled in about six months (;).

  3. First quarter dominated by three companies

    By Legal Complex's count, $2.34 billion was raised across 103 deals. Relativity's $720 million debt facility, Legora and Harvey together took 63% of it ().

  4. Crunchbase confirms the record and the slowdown

    Crunchbase put 2025 at a record $4.6 billion and 2026 so far at $2.2 billion. It also noted heavy acquisition activity by Legora and Harvey ().

  5. Harvey raises $550M at $15.5B

    Diffusion and Lightspeed co-led the round. It closed less than six months after Harvey raised $200 million at an $11 billion valuation (;).

  6. Legora's Series E

    Legora raised $550 million at an $8.8 billion post-money valuation, led by Atreides Management. New investors include Fidelity, Capital Group and J.P. Morgan (;).

Capabilities and access

  • Harvey Tenet (research preview, Aug 20, 2026): Harvey's first in-house model. It is Moonshot AI's open-weight Kimi K3, further trained by Harvey with Fireworks.
  • Harvey Legal Agent Benchmark (LAB), May 6, 2026: Open-source on GitHub. It has more than 1,250 tasks across 24 practice areas (Harvey).
  • Legora aOS: An "agentic operating system" for legal work.
Read the full section
  • Harvey Tenet (research preview, Aug 20, 2026): Harvey's first in-house model. It is Moonshot AI's open-weight Kimi K3, further trained by Harvey with Fireworks. Harvey's post does not say that Tenet's weights will be released (Harvey).
  • Harvey Legal Agent Benchmark (LAB), May 6, 2026: Open-source on GitHub. It has more than 1,250 tasks across 24 practice areas (Harvey).
  • Legora aOS: An "agentic operating system" for legal work. It includes add-ins for Word and Outlook, plus tools for review, workflows, legal research and intake (Legora).

Technical analysis for researchers and developers

  • How Tenet was trained: Reinforcement learning using the GSPO method with low-rank (LoRA) adapters, run asynchronously across about 1,750 simulated legal-task environments.
  • How LAB grades: Expert-written pass/fail checklists, with about 75,000 criteria in total.
  • Reproducibility problems: Tenet's results were scored by another AI model, Kimi 2.6, rather than human graders.
Read the full section
  • How Tenet was trained: Reinforcement learning using the GSPO method with low-rank (LoRA) adapters, run asynchronously across about 1,750 simulated legal-task environments. Harvey says it used about 150 Nvidia B300 GPUs for two months (Harvey).
  • How LAB grades: Expert-written pass/fail checklists, with about 75,000 criteria in total. A task counts only if every item passes, so there is no partial credit (Harvey).
  • Reproducibility problems: Tenet's results were scored by another AI model, Kimi 2.6, rather than human graders. Harvey built and evaluated its own benchmark, and Tenet's weights are not public. Outsiders can rerun LAB on other models but cannot check Tenet's scores.

Claims and evidence

  • Round sizes, valuations and lead investors were announced by the companies and repeated by independent outlets (LawNext; Tech.eu).
  • Revenue figures are self-reported. Legora says its annual recurring revenue is above $200 million (Legora).
  • It reports almost twice as many held-out LAB tasks completed as base Kimi K3 and about 90% lower cost per query (Harvey).
Read the full section
  • Round sizes, valuations and lead investors were announced by the companies and repeated by independent outlets (LawNext; Tech.eu). No regulatory filings confirming them were found. Harvey's $15.5 billion target was first reported in August, citing unnamed sources (SiliconANGLE).
  • Revenue figures are self-reported. Harvey cofounder Winston Weinberg says annual recurring revenue passed $400 million, and Harvey claims more than 3,000 customers (LawNext). Legora says its annual recurring revenue is above $200 million (Legora).
  • Tenet's results come only from Harvey. It reports almost twice as many held-out LAB tasks completed as base Kimi K3 and about 90% lower cost per query (Harvey). No independent evaluation was found.

Context and prior work

  • 2025 totals disagree: Crunchbase counts $4.6 billion; Legal Complex counts $5.99 billion (Crunchbase; Artificial Lawyer).
  • 2026 totals disagree too: Legal Complex's first quarter alone ($2.34 billion, including debt) is larger than Crunchbase's $2.2 billion for the year through August.
  • Legora's Series D is reported two ways: $550 million in the company's announcement and $600 million in Crunchbase's write-up.
Read the full section
  • 2025 totals disagree: Crunchbase counts $4.6 billion; Legal Complex counts $5.99 billion (Crunchbase; Artificial Lawyer).
  • 2026 totals disagree too: Legal Complex's first quarter alone ($2.34 billion, including debt) is larger than Crunchbase's $2.2 billion for the year through August. The two trackers clearly define a "deal" differently.
  • Legora's Series D is reported two ways: $550 million in the company's announcement and $600 million in Crunchbase's write-up.
  • Harvey in 2025: It raised four rounds that year, totalling $818 million (Artificial Lawyer).

Limitations, safety and contested findings

  • Funding outside the leaders: Robin AI ran into funding trouble. In 2025, total exit value fell 39% (Artificial Lawyer).
  • Is the growth sustainable? It contrasts Harvey's speed with Clio and Litera, which took 17 to 30 years to reach similar revenue.
  • Gaps in the Tenet report: Harvey's write-up does not cover failure modes. Harvey says no customer data was used in training (Harvey).
Read the full section
  • Funding outside the leaders: Robin AI ran into funding trouble. Dozens of companies funded between 2020 and 2023 have not raised again. In 2025, total exit value fell 39% (Artificial Lawyer).
  • Is the growth sustainable? The same article questions it. It contrasts Harvey's speed with Clio and Litera, which took 17 to 30 years to reach similar revenue.
  • Gaps in the Tenet report: Harvey's write-up does not cover failure modes. Harvey says no customer data was used in training (Harvey).

Business and practitioner implications

  • For buyers: The market is consolidating around Harvey and Legora, and both are buying smaller companies.
  • For other startups: Early-stage money is still available.
  • For developers: LAB is open-source, so it can be used to evaluate legal agents without depending on any vendor's own scores.
Read the full section
  • For buyers: The market is consolidating around Harvey and Legora, and both are buying smaller companies. Crunchbase counts at least three acquisitions by Harvey and five by Legora in 2026, while Wolters Kluwer bought Brightflag for $500 million (Crunchbase).
  • For other startups: Early-stage money is still available. In the first quarter, seed deals outnumbered growth deals for the first time since early 2024 (Artificial Lawyer).
  • For developers: LAB is open-source, so it can be used to evaluate legal agents without depending on any vendor's own scores.
FOLLOW THE EVIDENCE

The source trail.

Sources (13)
A LITTLE LESS NOISE. A LOT MORE CONTEXT.

Stay curious.
Follow the evidence.

Independent perspectives, the original sources, and room for the questions that don't have easy answers.

How we build the brief
Connect with us

Find us where you already read.