Skip to content
LegalizeAI

Legora Review: Pricing, Fit, and the Leya Rebrand

By Mark Fulton12 min read

Legora Review: Pricing, Fit, and the Leya Rebrand

Legora publishes no prices. There is no pricing page on legora.com, the /pricing path returns a 404, and every route ends at a demo request, which is why Legora is recorded as "pricing on request" in our directory listing. The more important fact is that the pricing model changed on June 23, 2026: Legora launched a top tier called Agent Pro and put it on consumption-based billing rather than a flat per-seat fee. That means most of the dollar figures circulating for Legora describe a seat price for a product tier that no longer maps cleanly to what a firm would buy today. And if you are searching for "Leya," you are searching for the same company: the Stockholm-founded platform rebranded from Leya to Legora in February 2025.

We do not sell Legora, we do not sell a Legora alternative, and no vendor pays to change how a listing is described here. That matters on this particular query, because almost every page carrying a Legora price is published by a company selling a competing legal AI product, and the figures on those pages cite each other rather than a source.

What is Legora (and why was it called Leya)?

Legora was founded in Stockholm in 2023 by Max Junestrand, Sigge Labor, and August Erséus, and went through Y Combinator's Winter 2024 batch under the name Leya. On February 19, 2025 the company rebranded to Legora and expanded its product line at the same time, adding agentic research, a Microsoft Word add-in, and collaboration features on top of its tabular review tool.

That rebrand is the single biggest source of confusion on this topic. A large share of the indexed coverage, forum threads, and comparison pages about "Leya legal AI" was written before February 2025 and describes an earlier version of the same product. Nothing was acquired, nothing was shut down, and no team changed hands. It is a name change, and we track it alongside the other renames and acquisitions in this market in our legal AI acquisitions tracker. If you find a 2024 review of Leya, read it as an early review of Legora, and discount the feature list accordingly.

The company has scaled hard since. Silicon Republic reported in March 2026 that Legora raised $550 million led by Accel at a $5.55 billion valuation, with the platform used across more than 800 law firms and in-house legal teams in over 50 markets, and headcount growing from roughly 40 to 400 people across Stockholm, London, New York, Denver, Sydney, and Bengaluru. Named customers in that reporting include White & Case, Cleary Gottlieb, Linklaters, Bird & Bird, Goodwin, Deloitte, and Dentons.

That customer list tells you more about pricing than any leaked number does. This is a platform sold to large firms with procurement functions.

What does the platform actually include?

Legora describes its architecture as an agentic operating system, and the surfaces a lawyer actually touches break down like this, based on the product pages published by the company:

  • Tabular review. The feature most people mean when they say "Legora." You load a document set, ask a column of questions, and get a structured grid back rather than a chat transcript. It is aimed at M&A due diligence and litigation document sets.
  • Agent and Agent Pro. The agentic layer that executes multi-step work end to end using firm context. Agent Pro is the higher tier introduced in June 2026, and it is the piece that moved to consumption billing.
  • Legal research. Jurisdiction-aware research with structured citations. Unlike CoCounsel, it is not grounded in a single publisher's database, which is a genuine tradeoff rather than a defect. You trade a known citation backbone for independence from one content vendor.
  • Word add-in. Drafting, redlining, and proofreading inside Microsoft Word, with saved prompts, customizable playbooks for standard documents like NDAs, and automated actions covering template fill, anonymization, writing cleanup, and translation via DeepL. There is a separate Outlook add-in.
  • Editor, Portal, Monitors, Lists, Workflows, and a mobile app. A drafting surface, a client-facing collaboration portal, regulatory change tracking, and workflow orchestration.

On the security posture that procurement will ask about, the company states SOC 2 Type II, ISO 27001, ISO 42001, GDPR, and HIPAA compliance, no model training on customer data, ethical walls with cross-matter isolation, and full audit trails.

One thing worth knowing before a demo: Legora has not appeared in the major independent benchmark of this category. The Vals Legal AI Report, published February 27, 2025 by Vals AI with Legaltech Hub, tested Harvey, CoCounsel 2.0, vLex's Vincent AI, and Vecflow's Oliver across seven legal tasks. Legora was not among the participants. That is not a criticism of the product. It does mean your own pilot is the only benchmark you will have, so design it properly.

What is known about Legora pricing?

Here is the verified picture, and then the reported one, kept strictly separate.

Verified. Legora publishes no rate card. On June 23, 2026 the company announced consumption-based pricing for Agent Pro: customers pay for the work Agent Pro delivers, and cost can be attributed to the matter or project that drove it. Existing Legora Agent customers were told the product stays available at no additional cost with no changes to their contract terms. Alongside it Legora shipped spend controls: real-time dashboards showing consumption by organization, user, or project, administrator-set spending thresholds, and notifications before a limit is reached. The Lawyer reported the same shift on June 23, 2026 under the headline "Firms face hike in AI cost base as Legora gears up for pricing switch".

Reported, and not repeated here as data. Search results for Legora pricing surface per-seat annual figures ranging from roughly $1,000 to $8,000, plus asserted minimum annual contract values around $30,000 and claims of steep negotiated discounts. Trace the citations and they collapse into a loop: the figures originate on marketing blogs run by companies selling competing legal AI, each citing another such blog, none citing a quote, a contract, or a named buyer. We are not going to launder those numbers by restating them. The same dynamic distorts the Harvey pricing and CoCounsel pricing conversations, and we have written about the pattern across the category in our legal AI pricing guide.

The structural point is more useful than any leaked figure anyway. A consumption model breaks the arithmetic most firms use to evaluate legal AI. Seats times rate times twelve stops being the budget line. What you need instead is a usage forecast, a cap, and a decision about who absorbs the variance.

That decision has professional consequences, not just financial ones. Writing in Legal Futures on July 10, 2026, general counsel Eduard Grigoryan argued that firms must settle whether AI consumption is a disbursement or overhead before passing any of it to clients, and that getting it wrong risks conduct and accounts problems under the Solicitors Regulation Authority's rules in England and Wales. His second point applies everywhere: once AI appears as a line item on an invoice, clients start associating the output with the vendor rather than with the firm. Whatever your jurisdiction, this belongs in the engagement letter before the first bill, not after it.

Who is Legora the right fit for?

Legora fits firms and in-house teams where the work is document-heavy and collaborative, and where the bottleneck is reviewing many documents against the same set of questions. Tabular review is the feature that wins demos, and it wins them for a reason: due diligence and litigation review are exactly the shape of problem a grid solves better than a chat window.

It also fits teams whose lawyers will not leave Microsoft Word. The Word add-in is a first-class surface here rather than an afterthought, which matters more than it sounds. Adoption failures in legal AI are usually workflow failures.

It fits less well in three situations. If you need published pricing to budget before you buy, you will spend weeks in a sales process to learn a number. If you need a single publisher's citation backbone as a matter of policy, a research product grounded in Westlaw or Shepard's is a closer match. And if you are a solo or small firm, you are not the buyer this motion is built around. Our Harvey alternatives piece sorts the field by how you can actually buy, and Paxton at $499 per user per month or $2,999 per year is the clearest example of a research and drafting tool you can price without a call.

How does it compare to Harvey and CoCounsel?

On capability the three overlap heavily, and we cover that ground in Harvey vs CoCounsel vs Legora. What genuinely separates them in 2026 is the buying process.

Worked example: a 90-lawyer firm running a three-way shortlist

Take a mid-size commercial firm, 90 lawyers, corporate and disputes practices, a two-person innovation committee, and a hard budget cycle ending in a quarter. Here is what the shortlist actually looks like from the procurement side.

Buying question Legora Harvey CoCounsel
Published price None. /pricing returns 404 None None
Entry point Demo request Demo request Demo request
Pricing model Agent Pro on consumption billing since June 23, 2026; lower tier stays fixed fee Negotiated, seat-based per reported figures Negotiated, commonly bundled with Westlaw and Practical Law
Content grounding Independent of any single publisher Independent, with Lexis content available via the LexisNexis alliance Westlaw and Practical Law
Native Word surface Yes, plus separate Outlook add-in Yes Yes
Independent benchmark data Not included in the Vals Legal AI Report Included Included
Budget shape to model Variable, needs a usage forecast and a cap Fixed annual, seat count is the lever Fixed annual, often folded into an existing research spend

Read the bottom two rows together. Legora is the only one of the three whose cost you cannot fully fix in advance, and the only one with no independent benchmark result to sanity-check the demo against. Neither is disqualifying. Both are things you have to actively manage, and neither shows up on a feature grid.

The demo questions to ask, in this order. These are the ones that change the number on the contract.

  1. Which tiers are consumption-priced today, which are fixed fee, and where exactly is the line between them?
  2. What unit is metered, and what does one unit of a realistic task cost? Ask them to price a specific job: 200 contracts, 12 questions per contract, in tabular review.
  3. What is the cap, who can raise it, and what happens when it is hit mid-matter? A hard stop and a silent overage are very different risks.
  4. Can consumption be attributed to a client matter number at the point of use, and can we export that for billing?
  5. What is the renewal formula, including any annual uplift, and does the consumption rate itself reprice at renewal?
  6. Which lawyers count as licensed users, can seats be reassigned mid-term, and do trainees, paralegals, and knowledge management staff need their own?
  7. What migrates if we leave: prompts, playbooks, workspaces, and the tabular review templates we build?
  8. Will you support a paid pilot on our own matters, scored by our lawyers, before we sign an annual term?

Run the same eight questions at all three vendors and write the answers into one page. The comparison that decides this is a procurement comparison, not a feature comparison, and it is the one most shortlists never write down. For a fuller version of that process, see how to evaluate legal AI.

Frequently asked questions

Is Legora the same company as Leya?

Yes. Leya rebranded to Legora on February 19, 2025. Same company, same founders, same product line, continuously developed since 2023. No acquisition, no merger, no change of ownership. Any review, benchmark, or forum thread written about "Leya" before that date is describing an earlier version of Legora, so treat the feature lists in that material as out of date rather than wrong.

How much does Legora cost?

There is no published price and no self-serve plan, so no honest answer exists outside a quote. What is verifiable: Legora's top tier, Agent Pro, moved to consumption-based billing on June 23, 2026, so cost scales with usage rather than sitting flat per seat, and the lighter Agent tier remained available to existing customers at no additional cost with unchanged contract terms. The per-seat annual figures circulating for Legora predate that change and trace back to uncited posts on competitors' marketing blogs. Ask for a metered price against a realistic workload and a hard spend cap, and treat any confident single number you read elsewhere as unsourced.

Does Legora have a Word add-in?

Yes. Legora ships a Microsoft Word add-in that brings drafting, redlining, and proofreading into the document, with saved prompts, customizable playbooks for standard agreements, and automated actions for template fill, anonymization, writing improvement, and translation powered by DeepL. There is a separate Outlook add-in for email workflows.

Is Legora available to US firms?

Yes. Legora opened a New York office in March 2025 and, per Silicon Republic's March 2026 reporting, operates from Stockholm, London, New York, Denver, Sydney, and Bengaluru, with further US expansion planned in Houston and Chicago and a target of 300 or more US employees by the end of 2026. Its named customer base includes US and transatlantic firms such as White & Case, Cleary Gottlieb, and Goodwin. Buying still runs through a sales process rather than self-serve signup.

Sources

Legora and 50 other tools are listed in our directory, free and verified, with pricing posture stated for each. If you are shortlisting in this space, start with the legal research category, where every enterprise platform is flagged as demo-gated or self-serve before you book a call. Listings are free, so submit a tool we are missing.