Patlytics vs Solve Intelligence

Patlytics and Solve Intelligence now describe themselves the same way — one AI platform for the whole patent lifecycle — and each raised a $40M Series B within five months of the other. The descriptions converged. The products did not.

Patlytics started in 2024 at the litigation end, where the deliverable is a claim chart backed by evidence of use, and worked backwards into drafting. Solve started in June 2023 inside the drafting editor, shipped Charts on 9 December 2025, and bought its litigation depth on 31 March 2026 with Munich-based Palito.ai. Each is strongest where it started, and the distance between them is about two years of iteration on that half of the lifecycle.

The routing rule is therefore a counting exercise on your own docket, not a feature audit: count the claim charts your team produced last quarter, then count the applications it filed. The bigger number names the vendor.

Where Patlytics wins

• Charting is its mature surface, not its newest one. Detection maps a patent against a product limitation by limitation, segments long limitations automatically so evidence attaches to the right phrase, and returns an editable chart with citation-backed snippets and read-strength ratings on each mapping. Portfolio screening runs the same pass across hundreds of patents against several products at once. Patlytics has shipped that deliverable since 2024; Solve’s Charts is nine months old.
• US litigation adoption is the deeper of the two. The vendor claims more than 40% of the Am Law 100, naming Quinn Emanuel, Latham & Watkins, McDermott Will & Emery, Susman Godfrey, and Foley & Lardner, with Rivian, Panasonic, Canon, Nissan, Sanofi, and Asahi Kasei on the corporate side. Unaudited vendor figures — read them as scale signals, not measurements — but for a procurement committee that indexes on peer adoption in US patent litigation, Solve does not answer the question as easily.
• It can be driven from outside the app. The MCP server at https://mcp.patlytics.ai/mcp speaks Streamable HTTP, authenticates against a Patlytics account, and exposes five read-only tools. In Claude Code that is one command:

claude mcp add --transport http patlytics https://mcp.patlytics.ai/mcp

Prior-art lookup then happens where the associate is already working. Solve publishes no MCP server and no public API, so nothing about it automates from outside the browser.

• ISO 42001 is certified rather than pending. Patlytics holds SOC 2 Type 2, ISO 27001, ISO 42001, and GDPR. ISO 42001 is the certificate an AI-governance questionnaire asks for by name, and Solve’s own site states its status two ways — listed with the certifications on one page, described as in progress on another.

Where Solve Intelligence wins

• Drafting is the load-bearing half of patent work, and Solve owns it. The editor behaves like a shared document with a chat panel beside it, and the patent-specific machinery lives on the same page instead of in an attachment workflow: figure generation, chemical structures and Markush claims, biological sequences, antecedent-basis review. The Review pass flags inconsistent terminology and antecedent-basis defects inline with one-click corrections — the errors a junior associate burns an afternoon on and an examiner finds anyway.
• Europe is a first-class docket, not an export market. The Palito.ai acquisition and the Munich office added German, EPO, and UPC validity workflows, German court procedure, case-law research, obviousness charting under 35 U.S.C. 103, and sentence-level citations. Processing region is selectable per customer between the US and EU. Patlytics has no equivalent EPO or UPC surface and no published regional-processing option, which turns an EU procurement review into a negotiation rather than a checkbox.
• The adoption curve is shorter. Prosecutors open a document editor without being retrained; a set of analysis modules needs someone to own rollout. Solve counts 700-plus IP teams across six continents, with Intel, Siemens, Amgen, DLA Piper, Finnegan, Wolf Greenfield, and Bardehle Pagenberg named (vendor figures, unaudited).
• House style is a product feature. Firm-specific know-how gets encoded as reusable AI styles, workflows, and templates, so the second application benefits from what the first one taught the platform.

Pricing reality

Neither vendor publishes a price. Both quote after a demo, and neither offers a free tier.

The two per-seat numbers in circulation are both inferences, and neither belongs in a budget. Solve’s approximately $775 per user per month — about $9,300 a year — was reverse-engineered from a single NAPP member benefit offering 15% off an annual subscription with the discount valued near $1,400. Patlytics has no vendor or customer figure at all; third-party aggregators put it around $800-2,000 per user per month by extrapolating from adjacent platforms. Taken at face value, the top of the Patlytics estimate is roughly 2.5x the inferred Solve seat. That is an estimate measured against an estimate: it orders the two, and it cannot price either.

The meter is the part you actually negotiate, and it is not the same meter. Patlytics prices seats plus analysis volume — detection runs and portfolio screens are the expensive unit, so a licensing program charting 200 patents a quarter buys a different contract than four litigators doing invalidity work. Solve prices seats plus drafting volume, and Charts launched as its own product, so confirm whether it is bundled or a separate SKU and whether EU regional processing carries a premium. On both order forms, itemize the entitlement, the overage rate above it, and the renewal uplift cap.

Both vendors’ ROI math converges on the same number from opposite ends. Patlytics cites $30,000-plus in attorney time saved per claim chart and an 80% reduction in project time; Solve cites 60-80% time savings on drafting and a 265% increase in average weekly actions per user since its Series A. Every one of those is vendor-supplied and unaudited. The only figure that survives a renewal conversation is yours: price one matter you have already charted by hand and ten applications you have already filed, then put both quotes against that same work.

Implementation effort

Solve is the lighter install. Onboarding is account provisioning plus the house template library that somebody on your side has to write. The cost sits at the edges: there is no Word add-in, so attorneys who refuse to leave Word turn the rollout into a fight you lose, and with no API or MCP server, drafts leave the platform as documents rather than as data.

Patlytics costs more to stand up. It is a set of modules plus an agent surface the vendor now counts at 200-plus Skills, up from 150-plus in August 2026 — a number that moves every quarter, which is itself the warning. Pin entitlements to a named release in the order form and refuse roadmap language. The MCP server is the compensating benefit, and its boundary is exact: the five tools find similar patents, get patent details, search non-patent literature, list your portfolios, and list patents in your vault. Infringement, invalidity, FTO, and drafting runs stay in the web app where the audit trail lives, so a Claude Code workflow designed to trigger analysis runs over MCP hits a wall.

Charts from either platform are drafts. A chart served on a bad mapping is signed by a named attorney, not by a vendor, so keep an explicit sign-off step in the workflow and log which limitations were AI-mapped — the review record is what survives a challenge. What a jurisdiction requires you to certify or disclose about AI-assisted filings varies by court and by office; check the applicable rules with your own counsel rather than with either vendor.

Neither product replaces an IPMS. Docketing, annuities, and IP spend management stay with Anaqua or its equivalent on both paths. Buy either one as the analysis layer over a portfolio system of record you already run.

Verdict

• Pick Patlytics when evidence-of-use, invalidity, FTO, or SEP work is the recurring cost center; when the docket is weighted toward US litigation and licensing; when procurement wants an ISO 42001 certificate in hand rather than in progress; or when you want prior-art lookup inside Claude Code or another MCP client.
• Pick Solve Intelligence when filing volume dominates — dozens of applications a quarter across your art units — and charting is the smaller half; when EPO and UPC matters are a real share of the docket; when the IP department is European and EU-resident processing is a procurement gate; or when adoption risk is the binding constraint and a document editor is the only surface your prosecutors will reliably open.
• Pick neither when your attorneys will not leave Microsoft Word: DeepIP drafts inside Word and connects to IPMS and docketing tools, with 400-plus law firms and corporate IP teams and $25M raised in March 2026. When the job is landscape analysis, search coverage, and a defensible data license rather than litigation work product, PatSnap, Clarivate Derwent, or Questel is the correct layer and neither of these two replaces it. When patents are one slice of a mixed docket and vendor consolidation beats best fit on any single workflow, Harvey or Legora is already on the firm’s invoice. And when the need is prior art plus an occasional chart rather than a platform, XLSCOUT quotes below both.

Default pick when you cannot decide: Solve Intelligence, on volume. Outside litigation-heavy and licensing practices, drafting and prosecution consume more attorney-hours per year than charting does, so the drafting-weighted platform touches more of the docket for the same seat — and its inferred seat price is the lower of the two. Buy it on the drafting case, then run Charts as a scoped POC against a matter you have already charted by hand and diff it limitation by limitation. If that diff comes back short, or if your quarterly chart count starts outrunning your filing count, that is the signal to add Patlytics for the litigation half rather than to switch. Quinn Emanuel appears on both vendors’ customer lists, which is the honest read on how this resolves at scale: the two coexist inside the same firm, and only the smaller team has to choose.