Mentat
Also known as: 멘타트, Mentat Legal
Early stage Korean legal agent that drafts litigation documents, complaints, answers, and preparatory briefs, from uploaded materials, with a full self drafting mode and a paragraph by paragraph co pilot.
Mentat is an early stage Korean legaltech startup building an AI agent that drafts litigation documents for lawyers. Named after the human computers of the novel Dune, it is positioned to reduce the burden of reviewing materials and drafting rather than to replace a lawyer's judgment. A Full Self Drafting mode generates drafts of complaints, answers, and preparatory briefs from uploaded case materials, while a co pilot mode lets a lawyer collaborate with the AI paragraph by paragraph. The company frames legal drafting as well suited to LLM based agents because it involves reviewing large volumes of data, extracting key points, and structuring logical arguments, and it reported that a private test with 60 legal experts cut drafting time by 60 to 95 percent. Founded in September 2025 and led by chief executive Kim Joo soo, Mentat raised a seed round of about 800 thousand dollars in 2026 and planned a launch that year with usage based plans for individual lawyers and small firms plus enterprise plans for larger firms.
Vendor details
Canonical URL
https://mentat.kr
Category
Enterprise operations agent
Funding status
Seed (about 800 thousand dollars, 1.1 billion won), 2026
Company status
private
Use cases & customers
Primary use cases
Target customers
Integrations
Drafts litigation documents from materials the lawyer uploads; system integrations, tool calling, and APIs are not documented.
Sources & related URLs
Related / legacy domains
Capability coverage
2.5 / 14 capabilities · 18%
| Integrations & Tool Callingdocuments are drafted from uploaded materials; system integrations and tool calling are not documented (venturesquare, wowtale) | Unable to verify |
|---|---|
| Workflow Orchestrationa full self drafting mode runs a multi step drafting workflow that reviews materials, extracts key points, and structures a draft; broader orchestration is not detailed (venturesquare) | Partial |
| Knowledge Grounding & RAGdrafts are grounded in the case materials the lawyer uploads, reviewing large volumes to extract key points; a persistent legal knowledge base is not detailed (venturesquare) | Partial |
| Human Oversight & Guardrailsa co pilot mode lets the lawyer refine drafts paragraph by paragraph, and the product is positioned to support rather than replace the lawyer's judgment (venturesquare, wowtale) | Full |
| Security, Identity & Governanceenterprise plans are positioned for clients with high security requirements, but specific security controls or certifications are not documented (venturesquare) | Unable to verify |
| Observability & Auditabilityobservability or audit logging is not documented (venturesquare) | Unable to verify |
| Memory & State Persistencememory or state persistence is not documented (venturesquare) | Unable to verify |
| Deployment & Data Residencydeployment and data residency options are not documented (venturesquare) | Unable to verify |
| Prebuilt Agents / Templates / Packsgenerates specific litigation document types including complaints, answers, and preparatory briefs as prebuilt drafting templates (venturesquare) | Partial |
| Triggers & Channel Coveragedrafting is initiated by the lawyer uploading materials; broader triggers or channels are not documented (venturesquare) | Unable to verify |
| Model Flexibility & Routingthe service is built on large language models, but model choice or routing is not documented (venturesquare) | Unable to verify |
| APIs / SDKs / MCP Extensibilityno public API, SDK, or MCP is documented (venturesquare) | Unable to verify |
| Testing, Debugging & Optimizationa private test with 60 legal experts showed drafting time reductions, but no in product testing or evaluation framework is documented (venturesquare, wowtale) | Unable to verify |
| Browser / Computer-useno browser or computer use capability documented (venturesquare) | Unable to verify |
Pricing
Usage based subscription plus enterprise plans announced; exact prices not yet public (launching 2026)
A usage based subscription for individual lawyers and small firms plus enterprise plans; exact rates are not yet public.
Cost watchouts
As a usage based model, cost scales with drafting volume; published rates are not yet available.
Variable cost rationale
A usage based model means cost scales with drafting volume; published rates are not yet available.
Sales call required
Mixed (some tiers require a call)
Free / trial
Newly launching in 2026; early access.
Lowest paid plan
Not yet public.
Key ambiguities
The service is newly launching, so a usage based model is announced but exact prices are not yet public.
Related vendors
- 11th Estate — Agentic platform whose AI engine scans markets, matches an…
- Akro AI — On premise operational intelligence that automates document heavy…
- Algebra AI — Delivery led provider that builds and runs human governed AI agents…
- Alloy.ai — Commerce intelligence system for consumer brands that unifies four…
- Apaleo — Open, API first hospitality property management platform pivoting…
- Apprentice.io — AI native manufacturing platform for pharma, biotech, and cell and…