Harvey
An AI startup focused on the legal industry, providing vertical AI solutions for law firms and corporate legal departments. Its offerings cover professional legal tasks including document review, case research, and contract drafting.
Core Facts
Timeline (last 90 days)
Anthropic has launched a library of legal-specific Artifacts tools for Cloud for Legal, which currently integrates more than 60 legal professional connectors and was developed with partners including Harvey and PwC.
Harvey announced the completion of a $500 million funding round.
Harvey's $500 million funding round points to the enterprise market, including the legal departments of large law firms, multinational corporations, and financial institutions, with the capital used to accelerate the capture of these seats.
What companies like Harvey sell is not just model capability but also the certainty that "someone is accountable if something goes wrong" (such as indemnity and SLAs).
A developer posted on Hacker News claiming to have built the same legal AI as Harvey and open-sourced it, and the post received 22 upvotes and 1 comment.
The researcher read public materials from 12 teams running agents in production environments, including Codex, Claude Code, Cursor, Devin, Stripe's Minions, Manus, Agentforce, Shopify Sidekick, Sierra, Intercom Fin, Harvey, and Hebbia.
Harvey keeps document state firmly outside the model.
Gemini officially claims to reach four times Astra's level on the Harvey legal Agent Benchmark.
Reducto's customers include Harvey, Scale, and one of the world's top five hedge funds.
Historically, Zapier, Airtable, and Notion took the general route and relied on template libraries and community ecosystems to lower the barrier to entry; Harvey (legal), Abridge (medical), and others relied on vertical depth to quickly build moats.
31 more timeline events
All Facts (20)
The API model identifier for GPT-6 Astra is gpt-6-astra.
90%VerifiedHarvey is an AI startup focused on the legal industry.
75%UnverifiedHarvey采用'协作优先'（collaborative by design）的工程哲学，从第一天就把协作当作核心需求来设计，而非先做好个人工具再打通
85%UnverifiedHarvey的核心产品是面向律师事务所和企业法务部门的AI助手，能够处理合同审查、法律研究、诉讼分析等专业任务
85%UnverifiedGPT-6 Astra is capable of generating legal documents with clearer structure and closer alignment to specific contexts.
70%UnverifiedIssues encountered by collaborative agents in practice include information overload, context pollution, and inadequate conflict-handling mechanisms caused by poorly designed collaboration workflows.
60%UnverifiedWhat companies like Harvey sell is not just model capability but also the certainty that "someone is accountable if something goes wrong" (such as indemnity and SLAs).
50%UnverifiedHarvey's $500 million funding round points to the enterprise market, including the legal departments of large law firms, multinational corporations, and financial institutions, with the capital used to accelerate the capture of these seats.
50%UnverifiedThe researcher read public materials from 12 teams running agents in production environments, including Codex, Claude Code, Cursor, Devin, Stripe's Minions, Manus, Agentforce, Shopify Sidekick, Sierra, Intercom Fin, Harvey, and Hebbia.
50%UnverifiedHarvey keeps document state firmly outside the model.
50%UnverifiedGemini officially claims to reach four times Astra's level on the Harvey legal Agent Benchmark.
50%UnverifiedReducto's customers include Harvey, Scale, and one of the world's top five hedge funds.
50%UnverifiedHistorically, Zapier, Airtable, and Notion took the general route and relied on template libraries and community ecosystems to lower the barrier to entry; Harvey (legal), Abridge (medical), and others relied on vertical depth to quickly build moats.
50%UnverifiedApplication-layer companies such as Harvey, Replit and Cursor have already achieved commercial success.
50%UnverifiedOn the Harvey Legal Agent benchmark, Gemini 4 Argon reaches 19.6%, versus a maximum of 6.7% for competitors.
50%UnverifiedHarvey LAB comes from within the industry rather than an academic institution, and its task design is closer to the real workflows of law firms.
50%UnverifiedHarvey LAB (Legal Agent Benchmark) simulates Agent tasks from real lawyer workflows, requiring the model to have comprehensive capabilities such as multi-step reasoning, document retrieval, clause comparison, and structured output.
50%UnverifiedThe background of OpShield's creation is that after Guardrails AI was acquired by Harvey, the market lacked an independent, self-hostable security alternative.
50%UnverifiedHarvey and van der Hoeven proposed an O(n log n) integer multiplication algorithm, which is considered to be asymptotically near-optimal.
50%UnverifiedThe experimental design was inspired by a workflow from legal AI company Harvey: a small company acquisition review.
50%